Taking a Trading System Live - futures io
futures io



Taking a Trading System Live


Discussion in Trading Journals

Updated
      Top Posters
    1. looks_one kevinkdog with 260 posts (717 thanks)
    2. looks_two Big Mike with 61 posts (97 thanks)
    3. looks_3 deaddog with 25 posts (24 thanks)
    4. looks_4 swz168 with 20 posts (40 thanks)
      Best Posters
    1. looks_one kevinkdog with 2.8 thanks per post
    2. looks_two swz168 with 2 thanks per post
    3. looks_3 rk142 with 2 thanks per post
    4. looks_4 Big Mike with 1.6 thanks per post
    1. trending_up 67,745 views
    2. thumb_up 1,131 thanks given
    3. group 97 followers
    1. forum 529 posts
    2. attach_file 189 attachments




Welcome to futures io: the largest futures trading community on the planet, with well over 125,000 members
  • Genuine reviews from real traders, not fake reviews from stealth vendors
  • Quality education from leading professional traders
  • We are a friendly, helpful, and positive community
  • We do not tolerate rude behavior, trolling, or vendors advertising in posts
  • We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community.  It's free and simple.

-- Big Mike, Site Administrator

(If you already have an account, login at the top of the page)

 
Search this Thread
 

Taking a Trading System Live

(login for full post details)
  #1 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

OK, if you'd read my other journal here at Big Mike's, you'll know (as of today) that I have attempted 2 TopStepTrader Combines in the past few months. I did not achieve the profit goal in either of them , but qualified for a rollover ("a do over") each time by finishing slightly profitable, and by meeting all other Combine criteria.

While not passing the Combine has been disappointing and more than a little embarrassing, it did have a silver lining...

When I developed the trading strategies for use in the Combine, I thought that they might be appropriate for my own trading, even if they weren't good enough for the Combine.

That was back in March 2013. Now, 5 months later, I have enough "real time" data to make the decision to go live.

And that is what this thread is about - once you've developed a strategy, and watched it in real time or sim traded it for a while, then what?


- When do you decide to go live, and what criteria should you use?

- How do you actually trade it? Automated, manual, etc.? Do you need a new account, or a new broker?

- How do you determine how much to fund the account with?

- How do you position size? Do you start small, or big? Do you have a position sizing scheme?

- If things go bad, when do you quit trading the system?

- How do you monitor it on an ongoing basis?

- How do you track mistakes?


In due time, I'll address all these questions and more in this thread, along with showing the ongoing performance. I'll also answer any questions you may have.

If you are looking for details of the trades my system takes, or a discussion of why I took certain trades, you have come to the wrong place - this thread is about the trading process, not the trades themselves.

I truly have no idea how this system will do over the next few months. My hope, as always, is that is will do great, but as with any strategy, I am always prepared to cut my losses and stop trading it if need be.

Comments and questions are appreciated!

Follow me on Twitter Started this thread Reply With Quote

Can you help answer these questions
from other members on futures io?
Trading floor pivot levels
Traders Hideout
Need help with strategy PriceActionSwingPro
NinjaTrader
Anybody uses Bookmap with Tradestation?
TradeStation
options data for chart
NinjaTrader
Multichart NET set signal parameters different than the …
MultiCharts
 
Best Threads (Most Thanked)
in the last 7 days on futures io
The Crude Dude Oil Trading System
77 thanks
Spoo-nalysis ES e-mini futures S&P 500
68 thanks
I finally blew up an account
62 thanks
FIO Journal Challenge - April 2021 w/Jigsaw Trading
38 thanks
The tiyfTradePlanFactory indicator
29 thanks
 
(login for full post details)
  #3 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Let's Start at The Beginning...

To avoid confusion later on, here are some basics you need to know:

Trading System Name: NGEC - this stands for "Not Good Enough for the Combine." Why such an odd name? Well, I had to call it something, and I always like to remind myself of my trading shortcomings. As much as I like this system, and as well as it may or may not do, it still wasn't good enough to pass the Combine. That to me is important to remember, as it helps keep me humble. I know quite a few very good traders, and also quite a few arrogant traders. Interestingly enough, though, I don;t know any arrogant AND very good traders.



2 Trading Strategies in NGEC:

Strategy #1 - Trades Overnight session, has high winning percentage, lots of little wins and an occasional big loser
Strategy #2 - Trades Day session, lower win percentage, primary profit generator

All strategies are out by 3 PM ET each trading day. Both strategies are independent, and I'll only be in one at any given time.



Market:

Euro Currency Futures (6E)




With the basics out of the way, my next post will be the "decision point" post. Do I go live, or not?

Follow me on Twitter Started this thread Reply With Quote
The following 5 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #4 (permalink)
 gulabv 
Dallas, TX
 
Experience: Beginner
Platform: Ninjatrader
Broker: Zen-Fire
Trading: ZN, 6E
 
gulabv's Avatar
 
Posts: 286 since May 2010
Thanks: 161 given, 167 received


kevinkdog View Post
Strategy #1 - Trades Overnight session, has high winning percentage, lots of little wins and an occasional big loser
Strategy #2 - Trades Day session, lower win percentage, primary profit generator

All strategies are out by 3 PM ET each trading day. Both strategies are independent, and I'll only be in one at any given time.



Market:

Euro Currency Futures (6E)




Hi @kevinkdog,

Thanks for starting this thread. Would you mind sharing the R:R ratio for the 2 strategies?

Reply With Quote
The following user says Thank You to gulabv for this post:
 
(login for full post details)
  #5 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


gulabv View Post
Hi @kevinkdog,

Thanks for starting this thread. Would you mind sharing the R:R ratio for the 2 strategies?


Here are the summary performance reports for each strategy. The Overnight Strategy #1 is the 74.86% winning strat.







Follow me on Twitter Started this thread Reply With Quote
The following 7 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #6 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

At this point, I have 3.5 or so years out walkforward backtest history (the key word being walkforward, if you don't understand the importance of this compared to a standard optimized backtest, please take some time to learn about walkforward testing). On top of that, I have about 5 months of "incubation" results - watching the NGEC system perform real time, with no changes to the original code (other than regularly scheduled re-optimizations).

Walkforward Results: July 2009 - Mar 2013
Incubation Results: March 2013 - August 2013


If the incubation results of the last 5 months "look" similar to the walkforward results, I should feel comfortable going live with the strategy.

Does it?


Here is how I determine if incubation and walkforward data "match." Keep in mind that I am not a statistician, so I tend to keep things simple, at the risk of not being 100% mathematically and scientifically rigorous. What I do passes a common sense test, though. I use 3 methods to check for a match:

1. Student's T Distribution test. This statistical test will tell you if 2 data groups (the walkforward results and the incubation results) are significantly different from each other.


You can pretty easily do this in Excel (with the TTest function), or you can use this handy calculator ( Data Entry: Student's t-test).

When I run this test, it tells me that there is a 56% chance of these distributions are not different.





2. Data Distribution Comparison. I create 2 histograms of the data. The first one is the actual data, and I lay the walkforward and the incubation results on top of each other. Do they look like they overlap? The second chart plots a theoretical normal curve histogram, based on the mean and standard deviation. Again, do these curves overlap?







3. Equity Curve Comparison. This is my favorite method, but is not very scientific or mathematical. I simply plot all the data and create an equity curve. When I do that, can I see where the walkforward ends, and incubation begins? If I can, that suggests something happened when incubation started, and that is usually a bad thing. If you wonder about this method, just create a strategy with optimized parameters, and then let it run live for a while. Most times, you'll notice a change in the curve.





In case you were wondering, here is the equity curve with the incubation period clearly marked.









CONCLUSION: Based on this analysis, I'd say the system is performing in Incubation the same as it performed in its walkforward test. In fact, Incubation is better than walkforward, which does concern me a bit (usually it is the other way around). But, it is close enough to give me confidence that I did not screw up during testing in development. It does not guarantee that when I go live, the system will be profitable - that is important to remember.


Next: I am ready to go live. This will start on Monday. Lots of questions still to address. My next post will address my "quitting point," since that helps me determine initial account size, initial position sizing, etc. I may not post this before Monday, but rest assured I have already completed the work (which is why I can trade on Monday).

Follow me on Twitter Started this thread Reply With Quote
The following 11 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #7 (permalink)
 deaddog 
Legendary Market Wizard
Prince George BC Canada
 
Experience: Advanced
Platform: IBs TWS
Broker: IB
Trading: Stocks
 
deaddog's Avatar
 
Posts: 879 since May 2013
Thanks: 135 given, 2,369 received

Looking at your stats, specifically annual returns, (2.62% & 6.67%) and question if the strategies are worth the time and effort?

Reply With Quote
 
(login for full post details)
  #8 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


deaddog View Post
Looking at your stats, specifically annual returns, (2.62% & 6.67%) and question if the strategies are worth the time and effort?

I agree, those percentages are very low. They are based on $100,000 account size, which is the Tradestation default for strategy Performance reports.

IMO, there are a lot of numbers in strategy reports that are useless, meaningless, or in this case, very misleading. "Account Size Required" is another misleading one.

You'll see, later on, that when the account is properly sized, and position sizing is added in the mix, the returns are more enticing.

Follow me on Twitter Started this thread Reply With Quote
The following 2 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #9 (permalink)
 deaddog 
Legendary Market Wizard
Prince George BC Canada
 
Experience: Advanced
Platform: IBs TWS
Broker: IB
Trading: Stocks
 
deaddog's Avatar
 
Posts: 879 since May 2013
Thanks: 135 given, 2,369 received


kevinkdog View Post
I agree, those percentages are very low. They are based on $100,000 account size, which is the Tradestation default for strategy Performance reports.

IMO, there are a lot of numbers in strategy reports that are useless, meaningless, or in this case, very misleading. "Account Size Required" is another misleading one.

You'll see, later on, that when the account is properly sized, and position sizing is added in the mix, the returns are more enticing.

OK I think I see. The return calculated is for one contract with a 100K account.

By trading more contracts or a smaller account the returns would increase.

Reply With Quote
The following 2 users say Thank You to deaddog for this post:
 
(login for full post details)
  #10 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Now that I have decided to start trading my NGEC ("Not Good Enough for the Combine") strategy live starting on Monday, I have to address the question that everyone likes to avoid when starting to trade a new strategy: If things go bad, when do I quit trading the strategy?

There are probably a million different conditions you could use as a basis for quitting a trading system. You could set a dollar amount, or possibly wait until you get a margin call, or wait until you run out of money. You could stop after X losers in a row, or X losing months. There is no "one" right or wrong answer.

But, there are 3 keys to setting a quitting point:

1. It should be based on the system you are trading. It makes no sense, for example, to quit after a 10% drawdown, if historically the system had 25% drawdowns before. This seems obvious, but you'd be amazed at how many people make arbitrary decisions like this, without taking the characteristics of the actual system into account.

2. Write it down. Refer to it often. Remember it. This may save you from disaster one day.

3. Follow it. If/when the written criteria is (unfortunately) met, STOP TRADING. This is a simple, but VERY difficult, step to follow.


I don't always use the same criteria for finding my quitting point, but here is how I am doing it for the NGEC system:

A. Look at walkforward history, and find the worst drawdown that occurred (daily basis). Multiply it by 1.5, since the worst drawdown is always in the future. For my system, that worst drawdown comes out to be $3,265. Multiply this by 1.5 to get $4,898.

B. Use Monte Carlo simulation to find the 95% level max drawdown. That means, in a years worth of trading, 95% of the time my maximum drawdown will be less than this amount. This turns out to be $5,082. (If I wanted to be more conservative, I could use the 99% level. That drawdown is $6,512.)

I should point out those drawdown figures assume 1 contract being traded the whole time. Yet, I will hopefully be trading more than 1 contract as time goes on. This could get confusing - my actual drawdown (with multiple contracts) could be a lot bigger than my drawdown limit (based on one contract). I just have to remember to calculate the 1 contract drawdown, and compare than to the $5K limit. This will be clearer later, when I set up my monitoring system.


Using results from points A and B above, I will take the average, and stop trading when the single contract drawdown reaches $5,000 (slightly rounded).

So, I have followed points 1 and 2 above. Time will tell if I follow point 3 -- I better if I need to!!!!!


Next post: How much money should I have to fund this trading system?

Follow me on Twitter Started this thread Reply With Quote
The following 16 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #11 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

At this point, I have figured out 1) that I will begin live trading my failed Combine NGEC system and 2) I will quit trading if my single contract drawdown hits $5,000. Now, I will determine account size.

This point is pretty important. Too little capital to start, and I may run out of $$ before the quitting point. On the other hand, too much capital and I will have a lower rate of return, as well as an inefficient allocation of capital.

Currently, the exchange initial margin on Euro currency is $2,750. So, add this to my "quitting point" drawdown, and I get $7,750. This is the minimum account size I should start with. This will allow me to trade up until my max drawdown is reached.


A couple of important points to consider:

1. I am assuming my broker requires exchange margin, even for day trading. If I had access to day trading rates, I could get by with less.

2. Margins can and do change. If the exchange required margin goes up, I may be forced to stop trading before hitting my quiiting point.

3. I am assuming that I am trading a single contract all the time.


As it turns out, I will want more than $7,750 in my account, for position sizing reasons. I am going to use $8,500, for reasons that will be revealed in the next few posts, which discuss position sizing.

Follow me on Twitter Started this thread Reply With Quote
The following 7 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #12 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

I did turn on the NGEC system (2 strategies) today, but so far no trades.

My plan will be to show account updates every week, or when something interesting happens.

I still have a bunch of topics to cover in the meantime, so I'll be posting those in some sort of logical order.

One interesting thing already occurred - I had to change the code a bit to handle order placement. It doesn't impact results, but it could have had I not fixed it. That is one of those "have to be live to witness it events." There is another one (rounding issue) that I also have to track down and debug. I will describe these issues as we progress.

Of course, if you have questions, criticisms or comments, please feel free to contribute. If this is to be a good forum thread, it has to be clear what I am doing and why. Politely challenge me, and I'll return the favor.

Follow me on Twitter Started this thread Reply With Quote
The following 4 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #13 (permalink)
 Big Mike 
Site Administrator
Swing Trader
Data Scientist & DevOps
Manta, Ecuador
 
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
 
Big Mike's Avatar
 
Posts: 50,077 since Jun 2009
Thanks: 32,543 given, 98,515 received

Your last combine seemed to suffer from discretionary decisions on sizing because the first couple of times you ended up with losing trades on increased size.

I am still trying to understand how you intend to approach sizing this time, and how you have backtested that portion of the equation.

Thx for taking the time to document your process.

Mike

We're here to help -- just ask

For the best trading education, watch our webinars
Searching for trading reviews? Review this list

Follow us on Twitter, YouTube, and Facebook

Support our community as an Elite Member:
https://futures.io/elite/

Visit other sites? Please spread the word about your experience with our community!
Follow me on Twitter Visit my futures io Trade Journal Reply With Quote
 
(login for full post details)
  #14 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Big Mike View Post
Your last combine seemed to suffer from discretionary decisions on sizing because the first couple of times you ended up with losing trades on increased size.

I am still trying to understand how you intend to approach sizing this time, and how you have backtested that portion of the equation.

Thx for taking the time to document your process.

Mike

Good question, and I will post answers in my next 2 posts about position sizing. It will definitely be simple, and easier to follow than my Combine position sizing. During the Combine, in a futile attempt to maximize profit, I generally sized each trade as large as I possibly could. That led to changing size practically every trade. It was unique position sizing based on the Combine requirements, and I would never recommend it for a normal account.


Preview answer: Fixed fractional sizing

Follow me on Twitter Started this thread Reply With Quote
The following 2 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #15 (permalink)
 deaddog 
Legendary Market Wizard
Prince George BC Canada
 
Experience: Advanced
Platform: IBs TWS
Broker: IB
Trading: Stocks
 
deaddog's Avatar
 
Posts: 879 since May 2013
Thanks: 135 given, 2,369 received


kevinkdog View Post



As it turns out, I will want more than $7,750 in my account, for position sizing reasons. I am going to use $8,500, for reasons that will be revealed in the next few posts, which discuss position sizing.

Some math to consider:

I took the annual return for 1 contract based on your reports and adjusted it for your position size.

Making the assumption that you have about a 50 – 50 split between the 2 systems with a 100K you will return $4780 per year. [(2890+6670)/2]

Based on your account size of $8500 it seems you are willing to risk $5000 to return $4780. A risk return of about 1 to 1 or no better than a coin flip.

Reply With Quote
The following user says Thank You to deaddog for this post:
 
(login for full post details)
  #16 (permalink)
 artemiso 
New York, NY
 
Experience: Beginner
Platform: Vanguard 401k
Broker: Yahoo Finance
Trading: Mutual funds
 
Posts: 1,129 since Jul 2012
Thanks: 765 given, 2,545 received


deaddog View Post
Some math to consider:

I took the annual return for 1 contract based on your reports and adjusted it for your position size.

Making the assumption that you have about a 50 – 50 split between the 2 systems with a 100K you will return $4780 per year. [(2890+6670)/2]

Based on your account size of $8500 it seems you are willing to risk $5000 to return $4780. A risk return of about 1 to 1 or no better than a coin flip.

Let's say I started a business with unlimited personal liability and a projected 5-year average income after tax of X=$25,000,000/year, using Y=$5,000,000 of my personal savings out of my total net worth of Z=$100,000,000.

You're saying that my risk-reward ratio is Z/X, which is worse than a coin flip. Right, that's technically true... but that's as good as saying that everyone has a risk-reward ratio of infinity since the worst case loss is dying from pulmonary embolism since sitting behind a computer screen to trade for too long incurs such a possibility.

Reply With Quote
The following 2 users say Thank You to artemiso for this post:
 
(login for full post details)
  #17 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


deaddog View Post
Some math to consider:

I took the annual return for 1 contract based on your reports and adjusted it for your position size.

Making the assumption that you have about a 50 – 50 split between the 2 systems with a 100K you will return $4780 per year. [(2890+6670)/2]

Based on your account size of $8500 it seems you are willing to risk $5000 to return $4780. A risk return of about 1 to 1 or no better than a coin flip.


Thanks for the analysis. Your assumptions, unfortunately, are off the mark, and that is leading you make some faulty conclusions.

1) The 2 strategies in the NGEC system are independent, and never trade at the same time. So, their net profit results are additive, not averaged. They are described here ( strategies #1 and #2).

2) The performance reports time period (used to calculate annual return and annual profit) include some "start up" data, which is used to calculate indicators, etc. So, the annual profit in the strategy report is off the mark. That is one reason I don;t rely on annual numbers given in perf report - they can be misleading. The reality is the first trade date is 7/28/09. Counting results from then until 8/16/13 yields 4.05 years. $49,317 (from the performance reports earlier in the thread) in profit during this period divided by 4.05 years is $12,177 average profit per year.

3) So, following your analysis, I'd be risking $5,000 to make $12,177, or a Reward:Risk of 2.44. But this isn't a true reflection of the Reward:Risk of this system, since I am not going to quit if/when I hit the average profit of $12,177 (Think of my system as a "trade." In that respect, the stop loss of this "trade" is fixed at $5,000, but the profit target of the "trade" is unlimited, which makes Reward:Risk unlimited). The reality is that I can risk $5,000 of capital, and possibly have a 1 year reward much greater than $12,177 ( per contract). It could also be a lot less , too.

The true reward:risk of this system can be calculated by Monte Carlo analysis, by taking the median annual return divided by the median maximum drawdown. I will be presenting this very soon, I promise. I have already completed the analysis before I started live trading - I just have to document it.

Thanks again for the question.

Follow me on Twitter Started this thread Reply With Quote
The following 4 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #18 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


artemiso View Post
Let's say I started a business with unlimited personal liability and a projected 5-year average income after tax of X=$25,000,000/year, using Y=$5,000,000 of my personal savings out of my total net worth of Z=$100,000,000.

You're saying that my risk-reward ratio is Z/X, which is worse than a coin flip. Right, that's technically true... but that's as good as saying that everyone has a risk-reward ratio of infinity since the worst case loss is dying from pulmonary embolism since sitting behind a computer screen to trade for too long incurs such a possibility.

Good explanation. I was having trouble thinking how to best explain it.

I wish I had your "Z" value of net worth!

Follow me on Twitter Started this thread Reply With Quote
The following 2 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #19 (permalink)
 DeadCatBounced 
Baltimore MD US
 
Experience: Intermediate
Platform: NinjaTrader
Trading: ES, NQ
 
DeadCatBounced's Avatar
 
Posts: 293 since Apr 2013
Thanks: 1,514 given, 725 received

Just wanted to say that I have a mechanical system that I am also working on, and after doing significant walk forward testing on it, it is now in its incubation period but I am also asking many of the questions you are going over in this thread, it is highly informative! Thanks for keeping this on the forum!

Visit my futures io Trade Journal Reply With Quote
The following user says Thank You to DeadCatBounced for this post:
 
(login for full post details)
  #20 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Trader Chris View Post
Just wanted to say that I have a mechanical system that I am also working on, and after doing significant walk forward testing on it, it is now in its incubation period but I am also asking many of the questions you are going over in this thread, it is highly informative! Thanks for keeping this on the forum!

Thanks. Please feel free to share whatever you are doing, and of course your comments, questions.

I should point out that the process I am laying out here is just how I do things. I can't say it is the best way, or the optimum way. I could be doing this all wrong, but then that could make for an informative thread, too.

It is different than I did things a few years ago, and may be different than I do things a few years from now.

I hope that by seeing how at least one person does it, people reading may get some do's and don'ts they can apply to their own trading.

Follow me on Twitter Started this thread Reply With Quote
 
(login for full post details)
  #21 (permalink)
 artemiso 
New York, NY
 
Experience: Beginner
Platform: Vanguard 401k
Broker: Yahoo Finance
Trading: Mutual funds
 
Posts: 1,129 since Jul 2012
Thanks: 765 given, 2,545 received


kevinkdog View Post
Good explanation. I was having trouble thinking how to best explain it.

I wish I had your "Z" value of net worth!

Well, but I don't think you want my risk of dying of pulmonary embolism or other trading-related health problems!

Reply With Quote
The following user says Thank You to artemiso for this post:
 
(login for full post details)
  #22 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


artemiso View Post
Well, but I don't think you want my risk of dying of pulmonary embolism or other trading-related health problems!

I probably already have your risk of dying, although from the heart, not the lungs. Heart disease runs in my family. My Dad had a double bypass in 1973, a quadruple bypass in 1983 and a triple bypass in the late '90s. He actually had a bypass of a bypass of a bypass.

The irony is he died with a relatively strong circulatory system. Cancer got him instead.

Stay Well!!

Follow me on Twitter Started this thread Reply With Quote
The following user says Thank You to kevinkdog for this post:
 
(login for full post details)
  #23 (permalink)
 artemiso 
New York, NY
 
Experience: Beginner
Platform: Vanguard 401k
Broker: Yahoo Finance
Trading: Mutual funds
 
Posts: 1,129 since Jul 2012
Thanks: 765 given, 2,545 received


kevinkdog View Post
I probably already have your risk of dying, although from the heart, not the lungs. Heart disease runs in my family. My Dad had a double bypass in 1973, a quadruple bypass in 1983 and a triple bypass in the late '90s. He actually had a bypass of a bypass of a bypass.

The irony is he died with a relatively strong circulatory system. Cancer got him instead.

Stay Well!!

Sorry to hear that.

I was talking to a friend a few weeks ago; someone who has a net worth above 10*Z and is about 70 years old.

He asked if I was eating well - I said I still cook, it's about $25/day but not too shabby. Hearing this, he said hey just let me know, we can go to O Ya (probably the most expensive restaurant in Boston - for contrast, we were having this discussion at the Bristol Lounge in Four Seasons and that's considered shabby) whenever I'm around.

Out of modesty, I quickly said no - but I regret it now - not because I missed on a chance to get a constant supply of $500 food, but rather because I realized his position: When you have a net worth of 10*Z, are about 70 years old, and still live on your own in a suite at the Four Seasons, an entire continent apart from your family, you'd pay anything to get genuine company for a night.

Reply With Quote
The following 2 users say Thank You to artemiso for this post:
 
(login for full post details)
  #24 (permalink)
 deaddog 
Legendary Market Wizard
Prince George BC Canada
 
Experience: Advanced
Platform: IBs TWS
Broker: IB
Trading: Stocks
 
deaddog's Avatar
 
Posts: 879 since May 2013
Thanks: 135 given, 2,369 received


kevinkdog View Post
Thanks for the analysis. Your assumptions, unfortunately, are off the mark, and that is leading you make some faulty conclusions.

Hey that’s the story of my life.

For the most part I like to keep my math simple so that I can understand it. Monte Carlo analysis is too far over my head for me to really understand it. I see any formula with more than one set of brackets and my eyes glaze over. Mention standard deviation, degrees of freedom or square root and I cover my ears and scream la la la.

So take it easy on a math challenged faulty conclusion maker.


kevinkdog View Post
1) The 2 strategies in the NGEC system are independent, and never trade at the same time. So, their net profit results are additive, not averaged. They are described here ( strategies #1 and #2).

How long do you plan to spend in front of a computer each day? Are the systems automated with defined exits set on entry?

Kevin Thanks for taking the time to answer.

Reply With Quote
The following user says Thank You to deaddog for this post:
 
(login for full post details)
  #25 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


artemiso View Post
Sorry to hear that.

I was talking to a friend a few weeks ago; someone who has a net worth above 10*Z and is about 70 years old.

He asked if I was eating well - I said I still cook, it's about $25/day but not too shabby. Hearing this, he said hey just let me know, we can go to O Ya (probably the most expensive restaurant in Boston - for contrast, we were having this discussion at the Bristol Lounge in Four Seasons and that's considered shabby) whenever I'm around.

Out of modesty, I quickly said no - but I regret it now - not because I missed on a chance to get a constant supply of $500 food, but rather because I realized his position: When you have a net worth of 10*Z, are about 70 years old, and still live on your own in a suite at the Four Seasons, an entire continent apart from your family, you'd pay anything to get genuine company for a night.

Spending time with a high net worth person, especially if he is a teacher, and if he is low on arrogance, sounds like a winning proposition - even if you go to McDonalds...

Follow me on Twitter Started this thread Reply With Quote
The following 4 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #26 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


deaddog View Post
Hey that’s the story of my life.

For the most part I like to keep my math simple so that I can understand it. Monte Carlo analysis is too far over my head for me to really understand it. I see any formula with more than one set of brackets and my eyes glaze over. Mention standard deviation, degrees of freedom or square root and I cover my ears and scream la la la.

So take it easy on a math challenged faulty conclusion maker.



How long do you plan to spend in front of a computer each day? Are the systems automated with defined exits set on entry?

Kevin Thanks for taking the time to answer.


No problem, I will try to keep the math simple. My Monte Carlo Excel spreadsheet requires no math skills, if you decide to use it...


According to my family, I spend most of my time in front of a computer anyhow, so adding the NGEC system to the mix is no big deal. In a later post, I will detail exactly how I am trading it (manual vs automated, attended vs unattended, VPS vs no VPS, etc). But I can tell you it is no real extra work.

Follow me on Twitter Started this thread Reply With Quote
The following user says Thank You to kevinkdog for this post:
 
(login for full post details)
  #27 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

If you have a good trading system, eventually you will want to start trading it with multiple contracts. There are a ton of position sizing schemes out there (Van Tharp wrote a huge book on the topic), so there is no right way to do it. There is no Holy Grail position sizing technique, though, where you get more reward for no extra risk. The simple way to put it is this way: if you trade more contracts, your reward goes up, but so does your risk.

So, here is what I am doing (at least for a while, hopefully once my size gets big I'll become less aggressive)...

As always, I start out with only 1 contract. Why? Going live almost always reveals issues that backtests, sim tests and incubation tests keep hidden. For example, if my strategy is automated, what if some quirk in my code sends multiple orders, or otherwise goofs up? Or, what if my slippage estimates are way off, and real world slippage actually makes my strategy unprofitable? My experience is that starting with 1 contract is the cheapest way to find out and correct any live trading issues.

A second reason I like starting with 1 contract is that I want to remain emotionally detached as much as possible from the strategy performance. 1 contract profit and loss swings won't impact me or my emotions. 10 contracts, right off the bat, would freak me out a bit - I'd be watching the system too much, and have too emotion invested in it. As profits (hopefully) accumulate, I can add contracts at a comfortable level, and not be emotionally disturbed by it. If things go really well, in 6 months or even a year or two, trading 10 contracts at a time with this proven system will seem natural to me.

Some people would say "if you have an edge, exploit it fast and furiously" by trading max size right off the bat. Edges disappear, so take advantage while it exists. That is a good argument, and I understand the concept. But, I also know how I best operate, and going "all in" at the start is not good psychologically for me.

A final reason I like starting with 1 contract is that I like the strategy to be self generating - profits will build the account, leading to more contracts, building it further, leading to even more contracts, etc. No profits mean no increase in size. That just makes sense to me - why allocate more money to a system that isn't generating profits?

One drawback to this approach is that it can take a long time to add that second contract. For example, if you decide to trade 1 contract for each $10K in your account, you will have to have 100% return to add 1 contract. Then you'll need another 50% gain to add a third contract. That can take a long time.

Some position sizing techniques take this into account (fixed ratio sizing comes to mind), but these approaches have some negative characteristics I do not like.

I get around this dilemma by sizing my account for roughly 1.5-2 contracts at the start. This would be equivalent to starting with $15K, in the example I gave just above. Then, I only need 50% gain to add a second contract. This still forces the system to perform well, but at the same time I get contract growth sooner. For me, it is a great tradeoff.

With all that in mind, here are the details:

For my NGEC system, I have decided to use fixed fractional sizing.

Ncontracts = X * Equity / BigLoss

where

Ncontracts = integer number of contracts, always round numbers down

X = fixed fraction, which I determined through Monte Carlo analysis. For this system, I am using 0.175 (I'll explain later how I got this value)

Equity = current equity value

BigLoss = Largest daily loss, $885 for my NGEC system


Using the above, I can create the following tables:





Note that my fixed fraction of 0.175 might seem awfully high. It may be for most people. I determine it based on risk of ruin, annual return and max drawdown. I use my Monte Carlo spreadsheet to calculate all that. That will be in Part 2 of "Position Sizing."

Follow me on Twitter Started this thread Reply With Quote
The following 9 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #28 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

In the last post on Position Sizing, I determined that using fixed fractional sizing with X=ff=.175 was my best alternative. Please realize that this is my personal preference, based on my personal goals and objectives, and that amount probably would not be right for you.

The question is, how did I arrive at this figure? Why not just trade 1 contract all the time, or use a fixed fractional value of .01 or .02 or .10 or .50? This post will address that.

To determine the position sizing scheme that is right for me, I use my handy dandy Monte Carlo simulator, the basic (1 contract) version of which you can download for free at the bottom of this post. For a given trading system, it will estimate the probabilities of risk of ruin, median max drawdown, median annual return for the first year of trading.

The baseline version of this calculator assumes 1 contract traded throughout the year, but the macro code can be edited to simulate different position sizing techniques, which is what I am doing here.

There are 4 performance numbers I look at:

Risk of Ruin - how likely am I to hit my defined lower cash balance. I want this number low.

Median Maximum Drawdown - I have roughly a 50% chance of hitting this maximum drawdown sometime during the year. That of course means my maximum drawdown could be much greater than this value, and it could be less. I want it as low as possible, with a personal upper limit I have determined from doing this exercise a bunch of times.

Annual Return - As with drawdown, I have a 50% chance of reaching this annual return, and it could be much higher or much lower. I want it as high as possible, but I have no lower limit threshold acceptable value (the only number I don't have criteria for).

Return/Drawdown Ratio - Astute readers will recognize this as the Calmar ratio, although true Calmar is calculated over 3 years, not just 1 year. I want this value as high as possible, and I have a lower limit for acceptability. (Just for reference, for professional CTAs a Calmar above 1 is considered pretty good. That means if you want 25% annual return, you have to be willing to accept a 25% drawdown).

Using that criteria, I can try a few different position sizing approaches, with some different parameter values. Note that this is not all encompassing - I have not analyzed many other potential position sizing schemes. Maybe one would work better than what I chose.

Before I reveal the results, I should mention that I played around with the starting balance a bit, although I am not showing those interim results. Basically, by adjusting the initial account size, I was striking a balance between having too much money in the account, being able to add on a second contract relatively quickly (without doubling my account size first) and keeping risk of ruin low. I eventually settled on $8500 as the start balance, a good trade off between all competing metrics.


Here are the results, with the one I chose highlighted ("ff" is the fixed fractional amount):



My selection meets all my criteria, and I am comfortable with it.


This position scheme is the right one for me, right now. BUT, depending on how things go, I might change it down the road, either to a different scheme altogether, or a smaller value of f (ie, I will become less aggressive as the account grows). I'll let the performance of the system dictate if and when that happens.

Attached Files
Register to download File Type: xls freemc.xls (462.0 KB, 67 views)
Follow me on Twitter Started this thread Reply With Quote
The following 10 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #29 (permalink)
 deaddog 
Legendary Market Wizard
Prince George BC Canada
 
Experience: Advanced
Platform: IBs TWS
Broker: IB
Trading: Stocks
 
deaddog's Avatar
 
Posts: 879 since May 2013
Thanks: 135 given, 2,369 received


kevinkdog View Post
:

Trading System Name: NGEC -


2 Trading Strategies in NGEC:

Strategy #1 - Trades Overnight session, has high winning percentage, lots of little wins and an occasional big loser
Strategy #2 - Trades Day session, lower win percentage, primary profit generator

All strategies are out by 3 PM ET each trading day. Both strategies are independent, and I'll only be in one at any given time.

Why are you combining the 2 strategies to make one system?

Why not trade each strategy as a separate system?

Reply With Quote
The following user says Thank You to deaddog for this post:
 
(login for full post details)
  #30 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


deaddog View Post
Why are you combining the 2 strategies to make one system?

Why not trade each strategy as a separate system?


Sorry for me being unclear. Maybe this will help.



The NGEC strategies:

Strategy #1 - 105 minute bars - runs from 6 PM ET to 7 AM ET

Strategy #2 - 60 minute bars - runs from 7 AM ET to 2 PM ET


So, each is independent, and does not know what the other one is doing. Each strategy operates in its own little time period, so there should never be an overlap in positions.


So, yes, they are being traded as separate systems, but within the same account. I do this because Strategy #1, while not providing a lot of profit (strat #2 is much better), takes advantage of strat #1 downtime and uses margin that would otherwise be stuck just in cash.

Also, I could apply separate position sizing schemes to both strategies, but in the interest of keeping things simple, I am just going with the same position size in both strategies. Probably not the optimum approach.


Hopefully that explains it better. If not, just let me know what is confusing, and I will try again.

Thanks!

Follow me on Twitter Started this thread Reply With Quote
The following user says Thank You to kevinkdog for this post:
 
(login for full post details)
  #31 (permalink)
 JGSmith 
Lübeck, Germany
 
Experience: Intermediate
Platform: NinjaTrader, MT4
Broker: FXCM, Interactive Brokers, Oanda
Trading: Forex
 
JGSmith's Avatar
 
Posts: 151 since Aug 2013
Thanks: 61 given, 55 received

I quite enjoy reading your posts. Very good and very thought out. As you mentioned, you are not saying it is the best way, but at least it is well thought out and therefore, if there is an issues, it should be easy to find where it went wrong based on how you are thinking through things.

Reply With Quote
 
(login for full post details)
  #32 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


JGSmith View Post
I quite enjoy reading your posts. Very good and very thought out. As you mentioned, you are not saying it is the best way, but at least it is well thought out and therefore, if there is an issues, it should be easy to find where it went wrong based on how you are thinking through things.

Thanks. Of course, if anyone thinks there is a better way for me to do any of us (better position sizing, more reliable and cheaper broker, etc), please feel free to contribute. Such suggestions will make this thread even better, as we test, evaluate and compare the alternatives.

Kevin

Follow me on Twitter Started this thread Reply With Quote
 
(login for full post details)
  #33 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

Whenever I start a new strategy live, it is always nice to start with a profit.

Unfortunately, that doesn't seem to happen too often for me! Maybe it is just me, or maybe I just remember the down days - I don't know.

Anyhow, 2 trades today, automation worked as expected (I'll go into detail on automation on a later post).

Here's the tracking graph I'll be using (I will show and describe all tools I will be using to track this strategy as time progresses).


Follow me on Twitter Started this thread Reply With Quote
The following 5 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #34 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

In a previous post, I presented some results with fixed fractional position sizing. Basically, the results say that in an "average" year (meaning, 50% of years will be worse, and 50% of years will be better), I expect to make $30,735 profit in that year, and hit a maximum drawdown of 38.1% sometime during the year.

That profit number seems a little too good to be true...and my motto is "if something seems too good to be true, it probably is." And that profit number does seem too good to be true. 362% rate of return in that first year. Seems high...

Remember, though, the actual rate of return could be just about anywhere on the spectrum. It is just that the 362% is the median value.

Below is a histogram of possible returns. It will be interesting to see if the Year 1 results is anywhere close to the median ending equity (black vertical line). If it is, I will be very happy. I'll still be happy if I even hit the 25% mark, which is a final equity of about $21000, which is 147% return for the year. Still in "too good to be true" region.


Follow me on Twitter Started this thread Reply With Quote
The following 2 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #35 (permalink)
 JGSmith 
Lübeck, Germany
 
Experience: Intermediate
Platform: NinjaTrader, MT4
Broker: FXCM, Interactive Brokers, Oanda
Trading: Forex
 
JGSmith's Avatar
 
Posts: 151 since Aug 2013
Thanks: 61 given, 55 received

Is that Monte Carlo? If so, then that is not quite what it is saying though it is a similar statement. That technically is saying that 50% of the time, final balance is that amount or less.

Reply With Quote
 
(login for full post details)
  #36 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


JGSmith View Post
Is that Monte Carlo? If so, then that is not quite what it is saying though it is a similar statement. That technically is saying that 50% of the time, final balance is that amount or less.

Yes, it is based on Monte Carlo. Can you explain the statement "that is not quite what it is saying..." - I am not sure I understand.

Are you referring to my statement: "Basically, the results say that in an "average" year (meaning, 50% of years will be worse, and 50% of years will be better), I expect to make $30,735 profit in that year, and hit a maximum drawdown of 38.1% sometime during the year." I know I am using the term "average" incorrectly in this sentence (hence the quotes around it), as it is really the median value - the true average would be much higher, due to the spike at the far right of the histogram.

Thanks in advance!

Follow me on Twitter Started this thread Reply With Quote
 
(login for full post details)
  #37 (permalink)
 Koepisch 
@ Germany
 
Experience: Beginner
Platform: NinjaTrader
Broker: Mirus Futures/Zen-Fire
Trading: FDAX
 
Posts: 524 since Nov 2011
Thanks: 383 given, 455 received

@kevinkdog: I'm an attentive follower of your thread and very appreciate the granularity of your analysis. In another thread we have discussed the measuring of drawdowns und i've wrote my primary style of drawdown analysis here . Recently i've found a term for my Kind of analysis "Start-Trade DrawDown". It's mentioned in Keith Fitschen newest book "Building Reliable Trading Systems" which i can strongly recommend.

With a default monte carlo analysis you ignore the result dependency of subsequent trades. I refer to "Winning strikes" and "Loosing strikes". It could be a huge difference in your money management if you have an strongly trade-to-trade correlated system. In one of my books i have an algorithm to test that. With an "Start-Trade DrawDown" analysis you can see what your largest drawdown would be, if you had started your system at every day in the last x years.

I'm courios to see the difference between both methods.

Koepisch

Edit: "result dependency of subsequent trades" means "serial correlation" - thanks kevinkdog, sometimes it's hard for a non native speaker

Reply With Quote
 
(login for full post details)
  #38 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Koepisch View Post
@kevinkdog: I'm an attentive follower of your thread and very appreciate the granularity of your analysis. In another thread we have discussed the measuring of drawdowns und i've wrote my primary style of drawdown analysis here . Recently i've found a term for my Kind of analysis "Start-Trade DrawDown". It's mentioned in Keith Fitschen newest book "Building Reliable Trading Systems" which i can strongly recommend.

With a default monte carlo analysis you ignore the result dependency of subsequent trades. I refer to "Winning strikes" and "Loosing strikes". It could be a huge difference in your money management if you have an strongly trade-to-trade correlated system. In one of my books i have an algorithm to test that. With an "Start-Trade DrawDown" analysis you can see what your largest drawdown would be, if you had started your system at every day in the last x years.

I'm courios to see the difference between both methods.

Koepisch


Yes, I agree with you:

If there is serial correlation in your data (where the result of Trade X is related to the result of Trade X-1), the Monte Carlo analysis is not 100% appropriate.

To test this, I usually use a Durbin-Watson statistic. I find, for most of my systems, they do not have this serial or auto correlation issue.

When I get a chance, I will try to run the "start trade drawdown" - I think it would be useful. If it looks a lot different than the Monte Carlo, then that will be VERY interesting!

Follow me on Twitter Started this thread Reply With Quote
The following user says Thank You to kevinkdog for this post:
 
(login for full post details)
  #39 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Koepisch View Post
@kevinkdog: I'm an attentive follower of your thread and very appreciate the granularity of your analysis. In another thread we have discussed the measuring of drawdowns und i've wrote my primary style of drawdown analysis here . Recently i've found a term for my Kind of analysis "Start-Trade DrawDown". It's mentioned in Keith Fitschen newest book "Building Reliable Trading Systems" which i can strongly recommend.

With a default monte carlo analysis you ignore the result dependency of subsequent trades. I refer to "Winning strikes" and "Loosing strikes". It could be a huge difference in your money management if you have an strongly trade-to-trade correlated system. In one of my books i have an algorithm to test that. With an "Start-Trade DrawDown" analysis you can see what your largest drawdown would be, if you had started your system at every day in the last x years.

I'm courios to see the difference between both methods.

Koepisch

Edit: "result dependency of subsequent trades" means "serial correlation" - thanks kevinkdog, sometimes it's hard for a non native speaker


NOTE: This post scratches the surface of a complicated issue...proceed at your own risk...


OK, I ran the analysis. For everyone reading this, the issue is that, depending on your trade data, Monte Carlo analysis is not always a correct tool to use.

So, to first test this (is Monte Carlo OK to use?), I run something called the Durvin-Watson statistic. It checks for positive and negative autocorrelation. I think you can google search to find this for an Excel spreadsheet, what it means, etc. I'm no mathematician, so I will defer from getting into detail on this calculation.


Suffice it to say, for my NGEC system, the analysis says there is no autocorrelation issue, so I can use Monte Carlo analysis.


So to answer Koepisch's excellent question - how does Monte Carlo analysis compare with the "Start Trade Drawdown" analysis he mentions a couple of posts ago?


Again, without getting into too much detail, here are the results:




With the uncertainty of this whole analysis, it says that the 2 methods give very similar values for drawdown. The Monte Carlo gives more "worst case" scenarios - higher drawdowns than the Start Trade method, and this shows up as a greater standard deviation.


Thanks, Koepisch, for helping me add to the analysis here. The "Start Trade Drawdown" approach is verty interesting and useful.

Follow me on Twitter Started this thread Reply With Quote
The following 3 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #40 (permalink)
 Koepisch 
@ Germany
 
Experience: Beginner
Platform: NinjaTrader
Broker: Mirus Futures/Zen-Fire
Trading: FDAX
 
Posts: 524 since Nov 2011
Thanks: 383 given, 455 received


kevinkdog View Post
NOTE: This post scratches the surface of a complicated issue...proceed at your own risk...


...

With the uncertainty of this whole analysis, it says that the 2 methods give very similar values for drawdown. The Monte Carlo gives more "worst case" scenarios - higher drawdowns than the Start Trade method, and this shows up as a greater standard deviation.

Thanks, Koepisch, for helping me add to the analysis here. The "Start Trade Drawdown" approach is verty interesting and useful.

Excellent and fast analysis, great @kevinkdog! Due to the similarity of the results i would trust even more in the system. Furthermore you can trust your monte carlo analysis related money managment decisions - because it's BACKED with practical "What-if" "Start Trade Drawdown" analysis.

Reply With Quote
The following user says Thank You to Koepisch for this post:
 
(login for full post details)
  #41 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Koepisch View Post
Excellent and fast analysis, great @kevinkdog! Due to the similarity of the results i would trust even more in the system. Furthermore you can trust your monte carlo analysis related money managment decisions - because it's BACKED with practical "What-if" "Start Trade Drawdown" analysis.

@Koepisch - I certainly appreciate you bringing this topic up! It is a worthwhile issue to discuss and understand.

Follow me on Twitter Started this thread Reply With Quote
 
(login for full post details)
  #42 (permalink)
 ahwii 
Austria, Vienna
 
Experience: Intermediate
Platform: CQG Trader, NT
Broker: Velocity
Trading: FDAX
 
Posts: 59 since Sep 2010
Thanks: 88 given, 55 received

Hello kevinkdog

I really love your thread and the way you determine your optimal position size. I am following a similar approach using Monte Carlo simulations and the maximum drawdown found via those simulations.

The 38% drawdown is something i would not want to see myself. Of course i am not suggesting that it is not right for you, i am merely curious as to why it is acceptable for you.

From your posts i assume you plan on rerunning the analysis with new data periodically to avoid "deviation from reality" in your position sizing. Is that correct?

Do you have a system stop loss over periods of time? Personally i am currently thinking about stopping the system when it is down a given amount (or % value) in X amount of time (1 week for example). I would the analyse the cause for this - maybe the system stopped to work, the markets changed fundamentally, flash crash, or maybe just some unprobable string of losses. If i find that i should trade the system despite this massive drawdown, i would start it again - with a system stop loss found by analysis including the new data.

Lastly, i recently read Van Tharp's book, and it really is a must read. Nassim Taleb's Fooled by Randomness is also good, although more on the philosophical side of things (but it nevertheless changed my approach to things).

What are your thoughts?

"I don't even see the code anymore. I just see blonde, brunette, redhead..." -Cipher, The Matrix
Reply With Quote
 
(login for full post details)
  #43 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


ahwii View Post
Hello kevinkdog

I really love your thread and the way you determine your optimal position size. I am following a similar approach using Monte Carlo simulations and the maximum drawdown found via those simulations.

The 38% drawdown is something i would not want to see myself. Of course i am not suggesting that it is not right for you, i am merely curious as to why it is acceptable for you.

From your posts i assume you plan on rerunning the analysis with new data periodically to avoid "deviation from reality" in your position sizing. Is that correct?

Do you have a system stop loss over periods of time? Personally i am currently thinking about stopping the system when it is down a given amount (or % value) in X amount of time (1 week for example). I would the analyse the cause for this - maybe the system stopped to work, the markets changed fundamentally, flash crash, or maybe just some unprobable string of losses. If i find that i should trade the system despite this massive drawdown, i would start it again - with a system stop loss found by analysis including the new data.

Lastly, i recently read Van Tharp's book, and it really is a must read. Nassim Taleb's Fooled by Randomness is also good, although more on the philosophical side of things (but it nevertheless changed my approach to things).

What are your thoughts?

Thanks for the kind comments, @ahwii. Here are some answers:

1. The 38% drawdown is OK to me, mainly because I trade a bunch of different strategies, and diversification will help mitigate this large drawdown. I agree it is big, though - maybe I should be taking less risk.

2. I might change my position sizing ONLY if the system does really well. In that (hopeful) case, I would scale back the risk I am taking. If it happens, I will detail it here.

3. My "system stop" point is a $5,000 drawdown, based on trading a single contract. The will be the maximum. I may cut it shorter, probably not, but if I do I'll detail my reasoning in this thread.

4. I like most of what Tharp and Taleb have to say. I don't agree with everything, but their works stimulate (and sometimes change) my thinking on things.

Follow me on Twitter Started this thread Reply With Quote
 
(login for full post details)
  #44 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

Here is summary info after the first week of trading...





Follow me on Twitter Started this thread Reply With Quote
The following 2 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #45 (permalink)
 Big Mike 
Site Administrator
Swing Trader
Data Scientist & DevOps
Manta, Ecuador
 
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
 
Big Mike's Avatar
 
Posts: 50,077 since Jun 2009
Thanks: 32,543 given, 98,515 received


kevinkdog View Post
NOTE: This post scratches the surface of a complicated issue...proceed at your own risk...


OK, I ran the analysis. For everyone reading this, the issue is that, depending on your trade data, Monte Carlo analysis is not always a correct tool to use.

So, to first test this (is Monte Carlo OK to use?), I run something called the Durvin-Watson statistic. It checks for positive and negative autocorrelation. I think you can google search to find this for an Excel spreadsheet, what it means, etc. I'm no mathematician, so I will defer from getting into detail on this calculation.


Suffice it to say, for my NGEC system, the analysis says there is no autocorrelation issue, so I can use Monte Carlo analysis.


So to answer Koepisch's excellent question - how does Monte Carlo analysis compare with the "Start Trade Drawdown" analysis he mentions a couple of posts ago?


Again, without getting into too much detail, here are the results:




With the uncertainty of this whole analysis, it says that the 2 methods give very similar values for drawdown. The Monte Carlo gives more "worst case" scenarios - higher drawdowns than the Start Trade method, and this shows up as a greater standard deviation.


Thanks, Koepisch, for helping me add to the analysis here. The "Start Trade Drawdown" approach is verty interesting and useful.

New to me.

I am always trying to look at correlation across a portfolio. Can you briefly tell me what inputs you are using and if you are generating this data yourself and piping it to Excel, or some other tool?

My hurdle has been trying to feed a dozen strategies (trade output) into Excel, having to do it manually is less than ideal.

Mike

We're here to help -- just ask

For the best trading education, watch our webinars
Searching for trading reviews? Review this list

Follow us on Twitter, YouTube, and Facebook

Support our community as an Elite Member:
https://futures.io/elite/

Visit other sites? Please spread the word about your experience with our community!
Follow me on Twitter Visit my futures io Trade Journal Reply With Quote
 
(login for full post details)
  #46 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Big Mike View Post
New to me.

I am always trying to look at correlation across a portfolio. Can you briefly tell me what inputs you are using and if you are generating this data yourself and piping it to Excel, or some other tool?

My hurdle has been trying to feed a dozen strategies (trade output) into Excel, having to do it manually is less than ideal.

Mike

I basically just input the trade by trade P/Ls. I do it manually.

When I do a portfolio of strategies, I'll copy/paste the daily P/L to Excel.

Once in Excel, I'd do a correlation test over the time period of interest.

You are right, doing it manually many times would be cumbersome.

Doesn't Ninja have some sort of porting to Excel feature, or DDE link? That might make it more automated for you.

Follow me on Twitter Started this thread Reply With Quote
 
(login for full post details)
  #47 (permalink)
 Big Mike 
Site Administrator
Swing Trader
Data Scientist & DevOps
Manta, Ecuador
 
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
 
Big Mike's Avatar
 
Posts: 50,077 since Jun 2009
Thanks: 32,543 given, 98,515 received


kevinkdog View Post
Doesn't Ninja have some sort of porting to Excel feature, or DDE link? That might make it more automated for you.

I have threads on this, somewhere... basically I found it easier to write my own output. Unfortunately, still terrible. I want actual portfolio management, backtesting and reporting built-in to the platform, but there is no demand for it apparently.

Mike

We're here to help -- just ask

For the best trading education, watch our webinars
Searching for trading reviews? Review this list

Follow us on Twitter, YouTube, and Facebook

Support our community as an Elite Member:
https://futures.io/elite/

Visit other sites? Please spread the word about your experience with our community!
Follow me on Twitter Visit my futures io Trade Journal Reply With Quote
 
(login for full post details)
  #48 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Big Mike View Post
I have threads on this, somewhere... basically I found it easier to write my own output. Unfortunately, still terrible. I want actual portfolio management, backtesting and reporting built-in to the platform, but there is no demand for it apparently.

Mike


I could guess at why there is no demand for it...the funny thing is, once you start trading multiple strategies, it takes on huge significance.

Follow me on Twitter Started this thread Reply With Quote
The following user says Thank You to kevinkdog for this post:
 
(login for full post details)
  #49 (permalink)
 deaddog 
Legendary Market Wizard
Prince George BC Canada
 
Experience: Advanced
Platform: IBs TWS
Broker: IB
Trading: Stocks
 
deaddog's Avatar
 
Posts: 879 since May 2013
Thanks: 135 given, 2,369 received


kevinkdog View Post
Here is summary info after the first week of trading...

Is that the summary for the week or for today.

It would be nice if you could present it in the same format s the daily combine report or at least on a trade by trade basis. It makes it easier for us novices to follow.

Reply With Quote
 
(login for full post details)
  #50 (permalink)
 ahwii 
Austria, Vienna
 
Experience: Intermediate
Platform: CQG Trader, NT
Broker: Velocity
Trading: FDAX
 
Posts: 59 since Sep 2010
Thanks: 88 given, 55 received


kevinkdog View Post
Thanks for the kind comments, @ahwii. Here are some answers:

1. The 38% drawdown is OK to me, mainly because I trade a bunch of different strategies, and diversification will help mitigate this large drawdown. I agree it is big, though - maybe I should be taking less risk.

2. I might change my position sizing ONLY if the system does really well. In that (hopeful) case, I would scale back the risk I am taking. If it happens, I will detail it here.

3. My "system stop" point is a $5,000 drawdown, based on trading a single contract. The will be the maximum. I may cut it shorter, probably not, but if I do I'll detail my reasoning in this thread.

4. I like most of what Tharp and Taleb have to say. I don't agree with everything, but their works stimulate (and sometimes change) my thinking on things.

Thanks for the response!
So the drawdown number is for one strategy, which makes the overall max drawdown smaller because the strategies are uncorrelated. Seems i have misunderstood your posts as i thought it was for the whole portfolio. Makes sense now, thanks for clearing that up.

Thanks for addressing the other points as well, now i better understand what you are doing. Looking forward to you posting here!

Have a nice weekend

"I don't even see the code anymore. I just see blonde, brunette, redhead..." -Cipher, The Matrix
Reply With Quote
 
(login for full post details)
  #51 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


ahwii View Post
Thanks for the response!
So the drawdown number is for one strategy, which makes the overall max drawdown smaller because the strategies are uncorrelated. Seems i have misunderstood your posts as i thought it was for the whole portfolio. Makes sense now, thanks for clearing that up.

Thanks for addressing the other points as well, now i better understand what you are doing. Looking forward to you posting here!

Have a nice weekend

Sorry, I think I have confused you. The NGEC System contains 2 uncorrelated strategies. The $5000 drawdown is a System stop (both strategies combined, trading 1 contract).

Follow me on Twitter Started this thread Reply With Quote
 
(login for full post details)
  #52 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


deaddog View Post
Is that the summary for the week or for today.

It would be nice if you could present it in the same format s the daily combine report or at least on a trade by trade basis. It makes it easier for us novices to follow.


That is the summary for the week.

I'll see what I can do. Thanks

Follow me on Twitter Started this thread Reply With Quote
 
(login for full post details)
  #53 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


deaddog View Post
Is that the summary for the week or for today.

It would be nice if you could present it in the same format s the daily combine report or at least on a trade by trade basis. It makes it easier for us novices to follow.


Here is the day by day accounting of P/L.


Follow me on Twitter Started this thread Reply With Quote
The following 3 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #54 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

I few posts ago @deaddog asked a question that got me thinking. I am using the same position sizing for the 2 strategies in my NGEC system - even though strategy #1 and strategy #2 are for the most part different. The only thing they have in common is the stop loss point, around $425 (34 ticks) per contract.

Deaddog got me thinking - maybe the position sizing should be different for the 2 strategies. Maybe that would improve overal performance metrics.

As with any trading idea or thought that pops up, I reserve judgment on it until I test and analyze it. The numbers will tell me if this is a good thing things to do or not. No emotion is the point, I suppose.

I won't bore you with the minutiae of my analysis, but I primarily looked at "trade 2 contracts of strategy #1 for every 1 contract of strategy #2."


Results

Current method
Acct Size: $8,500
Max DD: 38.1%
Annual Return: 362%


2/1 Sizing Method
Acct Size: $12,500
Max DD: 38.3% (same as current method)
Annual Return: 255%



Conclusion: I would need more money in the account to trade a 2 to 1 ratio, and my annual return would go down. So, it does not make sense.


Note: I performed a pretty simple analysis to conclude this. Really, what I should do is let the fixed fractional sizing for each strategy float, and find the optimum for each. I may do this in the future, but for right now I'm content to let it be.

Follow me on Twitter Started this thread Reply With Quote
The following user says Thank You to kevinkdog for this post:
 
(login for full post details)
  #55 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

At this point, I think I've described most of the details for trading my "Not Good Enough for the Combine" (NGEC) system. I know my starting account balance, my initial position size, my position sizing scheme, my quitting point and various other monetary details.

But what about accounting, and the trading broker?

I trade multiple systems live right now, and I use various accounts at various brokers. I do this for a few reasons. First, doing all the bookkeeping and accounting gets confusing when multiple systems are all lumped into the same account. More than once, "orphan" positions that I forget to close pop up. Having one system per trading account makes things a lot easier from a management standpoint.

The second reason I use multiple brokers is that sometimes brokers go belly up, or walk away with your money. I lost some money when PFG Best went out of business in 2012, when its founder revealed he had been forging bank statements for years. I still have only gotten back about 1/3 of my money, and I doubt I'll ever see it all. I'm no longer mad about it, but when I go back and look at the interview I did on Fox Business Channel right when the scandal broke, it is clear to me I was in a pissed off state of mind. I do not need that aggravation again. So, to me, spreading my risk around will keep me trading even if one broker fails.

The drawbacks to my approach are obvious. If I am worried about brokers failing, having more accounts with more brokers just increases the chances of me running into a bad broker, right? So, my approach may not be better than finding the best broker, and just putting all my eggs in that basket.

The second drawback is that using mulitple brokers leads to a less than optimal use of capital (margin) situation. This leads to less return, since more of my money is not put to use. That, to me, is an acceptable tradeoff.

So, for my NGEC system, I will open a new trading account. In addition to the reasons above, it will also help silence any naysayers (I can easily post redacted statements if need be).

After looking at my automation requirements, and the fact that all my code is written in Tradestation Easy Language, it makes the most sense to either use Tradestation as the broker, or a NinjaTrader broker. For the Ninja option, there is a neat little feature in Ninja that will take Tradestation generated signals, run them thru NinjaTrader and then send the signals onto a Ninja Broker. I've used it before with success, so it is a good option. The Tradestation option is the cleanest and easiest, of course.

When I looked into it, I found Tradestation had a new account promotion (even for existing customers like me), where basically I could get 2 months of commissions rebated back to me. That was a deal clincher for me. Down the road, I may look at a Ninja option, but for now Tradestation will be the way I go.


In the next post, I'll talk about automation and other issues.

Follow me on Twitter Started this thread Reply With Quote
The following 3 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #56 (permalink)
Pinot13
New York
 
 
Posts: 5 since Mar 2013
Thanks: 11 given, 3 received

Kevin:
Really enjoy your posts, please keep it up. During the last Combine, you had three strategies. What happen to Strategy 3?
Thanks

Reply With Quote
The following user says Thank You to Pinot13 for this post:
 
(login for full post details)
  #57 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Pinot13 View Post
Kevin:
Really enjoy your posts, please keep it up. During the last Combine, you had three strategies. What happen to Strategy 3?
Thanks

Thanks for the kind words. I hope I am giving info that someone could easily follow with their own strategy. I know a lot of these things I had to figure out by myself, and usually I learned them by making mistakes (which always seemed to cost me money!).

In Combine #1, I had 3 strategies. In Combine #2, I added a 4th strategy to the first 3 strategies (I trade #4 currently in another account).

Strategy #3 was really a "filler" strategy. I was worried about meeting the 20 trading days in 2 months rule, and I needed a strategy that gave me a handful (3-5) more trading days. I felt I needed it for the Combine, and it served its purpose.

Even though it was net positive over 105 trades in the past 4 years or so, it only averaged $28 per trade after commissions and slippage, and has actually lost more for the last 20 months (only about $1,500 loss over that time, but still...).

So, with all that in mind, I decided to ditch it. I really could have ditched strat #1 also, since strat #2 provides the bulk of the profits, but I decided to keep #1.

Follow me on Twitter Started this thread Reply With Quote
 
(login for full post details)
  #58 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

Here are some other topics I had to consider when taking my NGEC strategy live:


Backup Plans

In an ideal world, computers never crash, internet connections never go down, your broker always is up, etc. In the real world, lots of things can go wrong. Some things to consider - do you need any of these?:
  • Backup PC
  • Backup Data Storage (Offsite and Onsite)
  • Backup Internet Provider
  • Backup Power Supply
  • Backup Phone Line
  • Backup Broker
  • Backup Trading Desk

There is more, I know, but having backups (and possibly even backups for the backups) for everything on this list will get you a long way.


To Automated, or not to automate?

When I traded this system in the Combine, I had to have Tradestation alert me, after which I would manually places orders in the T4 platform. Over time, I missed a few trades, made a few mistakes, forgot to cancel open orders, etc. Overall, I don;t think these mistakes cost me any sim "money." In fact, they may have saved me a few bucks. But, that isn't the point. The point is I want to trade the system as I developed it. So, automation makes the most sense for me. Therefore, I will trade it automated.


Attended, Unattended?

Tradestation always warns its clients that "automated trading does not mean unattended trading." This is pretty sound advice, since issues pop up from time to time, Internet connections go down, orders get missed, etc. I plan on usually being around the PC when this strategy is running, so I'd say it qualifies as "semi attended." If my account grows, and my contract size gets significant, I'll revisit this approach.


VPS?

Many people use virtual private servers for their trading, to keep the downtime to a minimum, and their reliability high. I have never used one, but I'll keep the option open. In the last year, I've lost Internet connection only twice, and once was on a weekend. If my systems traded more than a few times a day, I'd probably do a VPS.


Where Orders Are Kept

I bring this up because many people don't know where their orders are. When your automated strategy fires an order, is it kept on your machine? On the broker's servers? At the exchange? Plus, different types of orders (limit, stop) may have different routing. For example, limit orders might be sent directly to the exchange, but stop orders may be held at the broker.

My point in bringing this up is that you should know where your orders are, and have plans in place in case something goes wrong. You might think you have an order at the exchange, but after your internet goes down and a fill was missed, you might realize it was really held on your PC. Not a good way to find out.




Next Post - how I am monitoring performance

Follow me on Twitter Started this thread Reply With Quote
The following 8 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #59 (permalink)
toucan94506
danville ca usa
 
 
Posts: 51 since Mar 2012
Thanks: 6 given, 41 received

Kevindog...


Your posts are fantastic... thanks for sharing your progress... a couple of questions:

when you say that you trade multiple systems live right now... do you mean that you already have multiple "automated" systems... or that you manually trade multiple systems.

I see that you too are using tradestation and programming in easy language. Are you developing systems using tradestation strategies or tradestation indicators. The reason that I ask, is that automated strategies can be hard to keep in sync with "live" orders. Writing similar code using an indicator and generating orders using something like OOEL or the older "placeorder" function will end up on the tradestation server or the exchange and you reduce sync problems.

I started out using strategies many years ago and switched to indicators when tradestation first came out with order functions like "placeorder". This way, i know when I send an order in, it will either reside on the tradestation computer or the exchange itself. This eliminates problems with failures of your computer or the internet. I manually choose when to turn on/off the automation during the day, then my code has entry logic and trade management/exits/stops are completely automated.

cheers and good luck with your strategy development.

toucan

Reply With Quote
 
(login for full post details)
  #60 (permalink)
 artemiso 
New York, NY
 
Experience: Beginner
Platform: Vanguard 401k
Broker: Yahoo Finance
Trading: Mutual funds
 
Posts: 1,129 since Jul 2012
Thanks: 765 given, 2,545 received

Edit: I think you can go a very long way if you spent some time working with other tools besides Tradestation/Excel.

Reply With Quote
The following user says Thank You to artemiso for this post:
 
(login for full post details)
  #61 (permalink)
 Nicolas11 
near Paris, France
 
Experience: Beginner
Platform: -
Trading: -
 
Nicolas11's Avatar
 
Posts: 1,071 since Aug 2011
Thanks: 2,232 given, 1,755 received

Hi @artemiso Which kind of tools do you have in mind?

Sorry for this intrusion, @kevinkdog , and congratulations for the quality of your journal!

Nicolas

Visit my futures io Trade Journal Reply With Quote
The following 2 users say Thank You to Nicolas11 for this post:
 
(login for full post details)
  #62 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Nicolas11 View Post
Hi @artemiso Which kind of tools do you have in mind?

Sorry for this intrusion, @kevinkdog , and congratulations for the quality of your journal!

Nicolas


Thanks. @artemiso is right. I know CTAs and hedge funds that use Tradestation, as do many retail people. But there are tons of other products much better, depending on your needs. Same goes for Excel.

For automation and backtesting specifically, I only do things I know my software can do correctly. Does that limit my systems, or make things more difficult for me? - Undoubtedly yes.

Follow me on Twitter Started this thread Reply With Quote
The following user says Thank You to kevinkdog for this post:
 
(login for full post details)
  #63 (permalink)
 Jura   is a Vendor
 
 
Jura's Avatar
 
Posts: 774 since Apr 2010
Thanks: 2,347 given, 688 received

Hi Kevin,

Interesting thread you got here. One question:

kevinkdog View Post
For automation and backtesting specifically, I only do things I know my software can do correctly. Does that limit my systems, or make things more difficult for me? - Undoubtedly yes.

If I may ask: why don't you expand your software and/or skill set? Do you believe it won't give much benefit or do you believe in 'don't change a winning team'? Or do you have more practical objections (i.e. no spare time)?

Reply With Quote
 
(login for full post details)
  #64 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Jura View Post
Hi Kevin,

Interesting thread you got here. One question:

If I may ask: why don't you expand your software and/or skill set? Do you believe it won't give much benefit or do you believe in 'don't change a winning team'? Or do you have more practical objections (i.e. no spare time)?

Good questions. It is probably a combination of things. Extra time is a big factor. Also, I've been able to solve my programming/strategy issues with software I have and know, although sometimes I know I am doing things the hard way (try trading exchange supported spreads with Tradestation. Oh, that's right, you can't with their main platform. So testing spread strats becomes a real chore.)

Some of it is just exposure to the "other side" - most of the people I talk to are retail, or have retail foundation (and have now moved on to CTAs, hedge funds), etc. I'm sure my tools would be different had I started in this by working at a hedge fund or bank.

I don't see myself running out of ideas to test and implement with the tools I have, but I think I know when that time might be. And if need be I'll do it.

I hope this answers your questions. It is a good topic to discuss.

Follow me on Twitter Started this thread Reply With Quote
The following 3 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #65 (permalink)
 artemiso 
New York, NY
 
Experience: Beginner
Platform: Vanguard 401k
Broker: Yahoo Finance
Trading: Mutual funds
 
Posts: 1,129 since Jul 2012
Thanks: 765 given, 2,545 received

In this position, I'd definitely spend some time with a typical programming language (Java, C#, C++, Python, MATLAB, R). I have my personal preferences among these languages, and others in mind, but any of them would suffice.

Reply With Quote
The following 3 users say Thank You to artemiso for this post:
 
(login for full post details)
  #66 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


artemiso View Post
But I personally see profit as a nonlinear reward for the value that you add: if you're using Tradestation, then you're consuming what little value that it provides.

Does that mean you believe going to higher value tools beyond Tradestation/NT can open the door to much higher reward (ie trading profits)?

Follow me on Twitter Started this thread Reply With Quote
 
(login for full post details)
  #67 (permalink)
 artemiso 
New York, NY
 
Experience: Beginner
Platform: Vanguard 401k
Broker: Yahoo Finance
Trading: Mutual funds
 
Posts: 1,129 since Jul 2012
Thanks: 765 given, 2,545 received


kevinkdog View Post
Does that mean you believe going to higher value tools beyond Tradestation/NT can open the door to much higher reward (ie trading profits)?

Tradestation/NT remind me of Notepad in this picture of learning curves:


Reply With Quote
The following 8 users say Thank You to artemiso for this post:
 
(login for full post details)
  #68 (permalink)
 swz168 
Nuremberg, Germany
 
Experience: None
Platform: MultiCharts
Trading: Fx
 
Posts: 49 since Jun 2010
Thanks: 32 given, 59 received


artemiso View Post
Yes. My answer sounds rather philosophical but anyway: Yes in that better tools allow you to create more value, and more efficiently. Tradestation really abstracts away all of the interesting work that you can create so the little value that you're creating is just in the strategy layer.

Thanks for your post. Now I'm really curious what tools there are which costs more than NT and what value they have. Would be glad to have more details.

I agree that tools can create more value/efficiency. But in general I wouldn't agree that the more expensive the better. Maybe I just missed out some great trading software which retail traders just don't know or even can't afford.

Reply With Quote
The following user says Thank You to swz168 for this post:
 
(login for full post details)
  #69 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


artemiso View Post
Whose idea was it that you must always:

1. Write strategies
2. Backtest.
3. If not promising, return to step 1.
4. Optimize/WFO etc.
5. Paper trade.

Great post! There is a lot to think about in it. My biggest gripe with retail software is that, in an attempt to make is super easy and user friendly, the software encourages and nutures a lot of bad habits, as well as creating pitfalls most people don't see until they lose money because of it (again, myself included).


Regarding the part of your post I quoted, I know that is how a lot of traders (at least retail) traders approach (myself included).

I'm curious, since you aren't from the retail world, how does your process differ, if you can share?

Follow me on Twitter Started this thread Reply With Quote
The following user says Thank You to kevinkdog for this post:
 
(login for full post details)
  #70 (permalink)
 swz168 
Nuremberg, Germany
 
Experience: None
Platform: MultiCharts
Trading: Fx
 
Posts: 49 since Jun 2010
Thanks: 32 given, 59 received

I think what makes the trader successful is not the tool, but the skills of the trader. The tool is just a "helper". He can be also very succesful with tools that cost nothing.

Reply With Quote
 
(login for full post details)
  #71 (permalink)
 artemiso 
New York, NY
 
Experience: Beginner
Platform: Vanguard 401k
Broker: Yahoo Finance
Trading: Mutual funds
 
Posts: 1,129 since Jul 2012
Thanks: 765 given, 2,545 received

I certainly agree. It's ultimately true, because regardless your toolset, there's eventually going to be a human interface somewhere in the stack.

Reply With Quote
The following user says Thank You to artemiso for this post:
 
(login for full post details)
  #72 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


swz168 View Post
I think what makes the trader successful is not the tool, but the skills of the trader. The tool is just a "helper". He can be also very succesful with tools that cost nothing.

I agree, but better tools can make for better trading.

I can hammer a nail in with my fist (theoretically), but it is much easier and effective if I buy and use a hammer. And, depending how much I need and use that hammer, maybe it is better for me to invest in a pneumatic hammer.

Follow me on Twitter Started this thread Reply With Quote
The following 2 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #73 (permalink)
 artemiso 
New York, NY
 
Experience: Beginner
Platform: Vanguard 401k
Broker: Yahoo Finance
Trading: Mutual funds
 
Posts: 1,129 since Jul 2012
Thanks: 765 given, 2,545 received

Edit: Typo.

Reply With Quote
 
(login for full post details)
  #74 (permalink)
 swz168 
Nuremberg, Germany
 
Experience: None
Platform: MultiCharts
Trading: Fx
 
Posts: 49 since Jun 2010
Thanks: 32 given, 59 received

Thanks for naming those tools. (I know you deleted the name after edit, so I let them here out too)


kevinkdog View Post
I agree, but better tools can make for better trading.
I can hammer a nail in with my fist (theoretically), but it is much easier and effective if I buy and use a hammer. And, depending how much I need and use that hammer, maybe it is better for me to invest in a pneumatic hammer.

Drastic example. 100% agree.

But we were comparing a 1000 $ tool to maybe 10000-100000 tool.
The tools mentioned by artemiso were all institutional tool. Nearly all retail wouldn't be able to apply those tool accordingly (e.g. under capitalization or lack of knowledge). And most of those functions, retail trader wouldn't even need them or wouldn't even know what to do with them.

Reply With Quote
 
(login for full post details)
  #75 (permalink)
 swz168 
Nuremberg, Germany
 
Experience: None
Platform: MultiCharts
Trading: Fx
 
Posts: 49 since Jun 2010
Thanks: 32 given, 59 received

Many traders say: Keep it simple. Often simple strategies tend to be stable and profitable in the long run.

Similar to trading software: Sometimes less is more. Why buy those expensive instutional software if you don't need most of those functions, I at least know for myself, that they won't add value to me, at least not now and not in the near future.

Edit:

Sorry for going off topic.

Discovered this thread today. Excellent, Kevin! Keep on the good work.

Reply With Quote
 
(login for full post details)
  #76 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


swz168 View Post
Many traders say: Keep it simple. Often simple strategies tend to be stable and profitable in the long run.

Similar to trading software: Sometimes less is more. Why buy those expensive instutional software if you don't need most of those functions, I at least know for myself, that they won't add value to me, at least not now and not in the near future.

Edit:

Sorry for going off topic.

Discovered this thread today. Excellent, Kevin! Keep on the good work.

Thank you for bringing the topic up. It is a good thing to discuss!

Follow me on Twitter Started this thread Reply With Quote
 
(login for full post details)
  #77 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

At this point, I have everything in place to begin the trading the system. If you think I've missed anything up to this point, please bring it up here.

Assuming things are in place, I'd like to share what I do once I go "live" with a strategy.

I use a variety of calculations, graphs, metrics, etc to help me monitor the strategy. I do not necessarily use them all on every strategy, but for this strategy, I will show all of them.

The first chart I use is what I call a "birdseye view" chart. It tells me, at a glance, how my strategy has performed historically, and in real time. To keep consistency, I use the same data source for all the data. In my case, it is the Trade List provided by Tradestation. You can get similar data from just about every trading platform out there. This is NOT actual real money results data, which will be covered in later charts and metrics.

The point of this chart will be to gauge the general overall effectiveness of real time performance (the green part of the curve on the right). Is the real time data consistent with the historical test and the incubation period data? If not, there may be something amiss. Maybe the strategy has stopped working correctly, due to market conditions for example. Or, maybe the assumptions made in the strategy about limit order fills are not realistic. This could especially be true in scalping type strategies, although really you may need actual real money results to check that. Bad assumptions or strategy development technique up front may not show up in historical or incubation tests, but they certainly will be revealed when real money is on the line.

So, I update this chart, and quickly review it. This gives me a general feeling whether or not my strategy is performing as expected. If it is, I can quickly move on to the next chart, which I'll discuss in a future post.

Also, if you don't know how to create an equity chart like I show below - don;t worry. I'll explain that, too - next post.


Follow me on Twitter Started this thread Reply With Quote
The following 4 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #78 (permalink)
 Jura   is a Vendor
 
 
Jura's Avatar
 
Posts: 774 since Apr 2010
Thanks: 2,347 given, 688 received


swz168 View Post
Thanks for your post. (...)


kevinkdog View Post
Great post! There is a lot to think about in it. (...)


swz168 View Post
Thanks for naming those tools. (I know you deleted the name after edit, so I let them here out too)
(...)

While I do understand why Artemiso edits his posts, I do find it somewhat annoying that in this case I missed the last page due to being in a different time zone. Normally, I wouldn't mind, but since this is an interesting discussion so I wonder what 'gold nuggets' I missed.

Did someone made a copy of the last page of the discussion (then send me a copy, if you wish), or am I now implicitly asking for copyright infringement?

Edit: PM received. Thanks.

Reply With Quote
 
(login for full post details)
  #79 (permalink)
 swz168 
Nuremberg, Germany
 
Experience: None
Platform: MultiCharts
Trading: Fx
 
Posts: 49 since Jun 2010
Thanks: 32 given, 59 received


kevinkdog View Post
At this point, I have everything in place to begin the trading the system. If you think I've missed anything up to this point, please bring it up here.

What timeframe are you trading?

Yes I think you have missed something:

I saw in previous posts that you are using Monte Carlo Simulation. The way you applied it, it is the most simple version. Because you "just" accept that normal distributions applies to your strategy, so the risk you calculated is not "true". I think you know this already.

Part of my full time is also risk management (not related to trading). There I also model risk and use Monte Carlo Simulation. Since normal distribution often doesn't apply for business related situations, I apply real distribution (and not normal) derived from the history. That will give you a more exact picture of your risk.

A powerful tool to do that is Palisade @risk. ( Palisade Corporation: Maker of Risk & Decision Analysis Software using Monte Carlo Simulation). It is an MS Excel Addon. The Industrial Version costs GBP £1,550.00. No free lifetime upgrades like NT or MultiCharts! But still affordable for retail traders.

What I like about @risk is the easiness of fitting distributions. Alone that is worth the money if you really want to apply "correctly" monte carlo simulation.

Reply With Quote
The following 3 users say Thank You to swz168 for this post:
 
(login for full post details)
  #80 (permalink)
 DarkPoolTrading 
PTA, Gauteng
 
Experience: Advanced
Platform: Self built + Sierra + TWS
Trading: Stocks and Options
 
DarkPoolTrading's Avatar
 
Posts: 1,036 since May 2012
Thanks: 1,244 given, 1,321 received

Just to add my 2c to the discussion on tools and software. I think what Kevin shows is that often (usually?) process is more important than tools.

Kevin has proved himself to be a world class trader, yet for the most part his tools are limited to Excel and Tradestation (correct me if im wrong?). However in most of Kevin's threads what is glaringly obvious is that he has a process that he has developed himself and follows with complete discipline.

Focus on the process of trading well and the rest will fall into place. All the fancy tools and software in the world wont make a bad trader profitable. But,...on the other hand good software and tools can aid an already process oriented and successful trader.

Diversification is the only free lunch
Follow me on Twitter Reply With Quote
The following 6 users say Thank You to DarkPoolTrading for this post:
 
(login for full post details)
  #81 (permalink)
 swz168 
Nuremberg, Germany
 
Experience: None
Platform: MultiCharts
Trading: Fx
 
Posts: 49 since Jun 2010
Thanks: 32 given, 59 received


DarkPoolTrading View Post
Just to add my 2c to the discussion on tools and software. I think what Kevin shows is that often (usually?) process is more important than tools.

Yes, I also think the process of designing a strategy is very important. E.g. avoid curve fitting. I try to have as few rules as possible and I always focus on the KISS rule.

But for the statisticial metrics/reports of my strategies, I also use more complex models to get more insights about my risks.

For me, software tools are "helpers". I want to be able to focus on strategy design and development. I don't want to program the broker interface, the charts etc. For these aspects, the tools save me months and years of time, and for that I also pay up to 2000$ (and not more!) for one software tool.

MultiCharts and @risk help me to save so much time and allowed me to implement my strategies. Without them, I would have a process plan and concept, but wouldn't be able to apply them.

Reply With Quote
The following 2 users say Thank You to swz168 for this post:
 
(login for full post details)
  #82 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


swz168 View Post
What timeframe are you trading?

The overnight strategy uses 105 minute bars, and the US day session uses 60 minute bars.


swz168 View Post
Yes I think you have missed something:

I saw in previous posts that you are using Monte Carlo Simulation. The way you applied it, it is the most simple version. Because you "just" accept that normal distributions applies to your strategy, so the risk you calculated is not "true". I think you know this already.

Yes, you are absolutely correct that I use a quite simple version of Monte Carlo. You are also correct that the trades in my system do not have a normal distrubtion:




BUT, and here is where I may be wrong in my thinking, I am not making any assumptions about data normality regarding the input data or the output data.

Or am I?

Maybe you can help me answer that...


INPUT DATA:

For the input data (the daily trades results), as shown above, the data is not normal. But, when I run the simulation, I only use the actual data as possible daily results. I do not assume the daily results could be different than what I input into the spreadsheet. I can see the value of that (why do I assume future values of daily results only be exactly what past daily results were? that is a very limiting assumption), and if I wanted to create a Monte Carlo model like that, then I'd have to assume sort of of distribution.

Example: I know the mean and standard deviation for my input daily results data. I could build a Monte Carlo simulator that would grab a random number from a normal curve generated from the aforementioned mean and standard deviation. If I did that, I would agree it would be incorrect - I would be assuming the data is normal, yet it clearly is not.


Can you comment on my use of input data? I do not think I am using a normal distribution anywhere. Am I making an assumption about data normality somehow that I do not realize?




OUTPUT DATA:




For the output data (the return and max drawdown over 1 year of simulated trades), the data is decidely not normal also.

When I analyze this data, I use percentile functions. I look at things like "what is the probability that my return will be greater than X?" I use the non-normal distribution above to answer the query.

Can you comment on my use of output data? I do not think I am using a normal distribution anywhere. Am I making an assumption about data normality somehow that I do not realize?



Looking back through the thread, I can see only one area where I assumed a normal distribution, although that was mainly for a visual look at the data, not for any Monte Carlo analysis.



I sincerely appreciate any comments in advance!

Follow me on Twitter Started this thread Reply With Quote
The following 3 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #83 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


DarkPoolTrading View Post
Just to add my 2c to the discussion on tools and software. I think what Kevin shows is that often (usually?) process is more important than tools.

I agree that the process is key. If you have no process, you have no repeatability, and that makes it really hard to create consistent systems.

I always wonder if my process is "right" or if there are better ways. @artesimo alluded to a different development process, and I hope he expands on how he does things.


DarkPoolTrading View Post
Kevin has proved himself to be a world class trader, yet for the most part his tools are limited to Excel and Tradestation (correct me if im wrong?). However in most of Kevin's threads what is glaringly obvious is that he has a process that he has developed himself and follows with complete discipline.

I would dispute the "world class" moniker - I am just a retail guy trying to survive every day. "World class" is a goal worth striving for, of course!


DarkPoolTrading View Post
Focus on the process of trading well and the rest will fall into place. All the fancy tools and software in the world wont make a bad trader profitable. But,...on the other hand good software and tools can aid an already process oriented and successful trader.

I agree!

Thanks for the post!

Follow me on Twitter Started this thread Reply With Quote
The following 3 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #84 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


swz168 View Post
A powerful tool to do that is Palisade @risk. ( Palisade Corporation: Maker of Risk & Decision Analysis Software using Monte Carlo Simulation). It is an MS Excel Addon. The Industrial Version costs GBP £1,550.00. No free lifetime upgrades like NT or MultiCharts! But still affordable for retail traders.

@swz168 -

I have no doubt this product (priced at $2,195 for 1 year) would provide a wealth of useful info and statistics. Back when I was in Quality Assurance management in aerospace, I recall one of my statisticians using this software for his work. He was savant like (and now works at the world renowned Cleveland Clinic).

If I gave you the raw data I use, and if we agree on 1 or more metrics to compare, would you be willing to run my data through your model?

I think it would be interesting to see the results - how a simple spreadsheet differs from a professional package.

Also, if I am presenting incorrect or misleading results, it would give me a solid reason to correct them.

Please PM me if you are willing, and we can proceed. I think this would be really interesting.

Kevin

Follow me on Twitter Started this thread Reply With Quote
The following user says Thank You to kevinkdog for this post:
 
(login for full post details)
  #85 (permalink)
 swz168 
Nuremberg, Germany
 
Experience: None
Platform: MultiCharts
Trading: Fx
 
Posts: 49 since Jun 2010
Thanks: 32 given, 59 received


kevinkdog View Post
[B]BUT, and here is where I may be wrong in my thinking, I am not making any assumptions about data normality regarding the input data or the output data.

Or am I?

Ok, my bad. While I quickly went through this thread, I read "z-Value". So I thought you used normal distribution. I checked your VBA code now. Indeed, you are not making the normal distribution assumption. You are using the real historical discrete distribution.

Now, if we assume your historical results have predictive power for the future, then I would still prefer to use a fitted distribution, because other outcomes of profit/and loss is also possible like in the real world. Also with discrete distribution, the results can be misleading, depending how much trades you have done. With less trades, the results will be more extreme if you have several outliers in your discrete distribution in your Monte Carlo Simulation.

As for the Random function in Excel, at least in academic circle, it is wideley criticized (wrong implementation). At least up to Excel 2007. Google will tell you more. I don't know about newer versions. don't know whether the quality of that function is good enough for trading.

By the way, how much Iteration do you use? 2500? This value is quite low, because with each simulation, you get values which differs maybe more than you like. I use 100 000 iterations. With higher values, I don't see significant difference.

I'm also curious about the difference in result with your approach and mine. So just provide me with your raw data.
I will try to do it next week. (Will PM you later). Mainly I want to compare end value of equity/value at risk (x%-percentile values, value at risk for y trades) and expectancy. Currently, I'm building up my risk modelling with @risk for trading.

As for @risk: the price includes a nonetime limited version. the one year maintenance means: you also get new main upgrades (not only minor bugfix for the version you bought) within one year free. This is what I remember.

Reply With Quote
The following 3 users say Thank You to swz168 for this post:
 
(login for full post details)
  #86 (permalink)
 swz168 
Nuremberg, Germany
 
Experience: None
Platform: MultiCharts
Trading: Fx
 
Posts: 49 since Jun 2010
Thanks: 32 given, 59 received


kevinkdog View Post

OUTPUT DATA:



I sincerely appreciate any comments in advance!

When I had a crash course in Monte Carlo simulation, I learned from a professor that the distribution of profit/loss is not normal distributed in the financial market. But the outcome (equity) is nearly normal distributed. So I'm wondering about your result.

Well, our comparison maybe will shed more lights into this issue.

Reply With Quote
The following user says Thank You to swz168 for this post:
 
(login for full post details)
  #87 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


swz168 View Post
Ok, my bad. While I quickly went through this thread, I read "z-Value". So I thought you used normal distribution. I checked your VBA code now. Indeed, you are not making the normal distribution assumption. You are using the real historical discrete distribution.

Good to hear. It is always nice to have an expert (I consider you one, as you analyze risk as a profession) check my work.



swz168 View Post
Now, if we assume your historical results have predictive power for the future, then I would still prefer to use a fitted distribution, because other outcomes of profit/and loss is also possible like in the real world. Also with discrete distribution, the results can be misleading, depending how much trades you have done. With less trades, the results will be more extreme if you have several outliers in your discrete distribution in your Monte Carlo Simulation.

I agree. This is one of those tradeoffs I made - complexity (and more power) versus simplicity. I chose simplicity, but I give up all the things you mention. It could very well be that my approach is too simple.



swz168 View Post
As for the Random function in Excel, at least in academic circle, it is wideley criticized (wrong implementation). At least up to Excel 2007. Google will tell you more. I don't know about newer versions. don't know whether the quality of that function is good enough for trading.

I agree. I don't know the impact, if any. I've tried some different random number generators over the years, and never noticed a big difference. But I realize random number generators are truly not 100% random. I might try a later version of Excel, to see if there is a measurable(to me) difference.



swz168 View Post
By the way, how much Iteration do you use? 2500? This value is quite low, because with each simulation, you get values which differs maybe more than you like. I use 100 000 iterations. With higher values, I don't see significant difference.

I agree again. I picked 2,500 iterations because the macro code (the way I wrote it) is terribly slow.


swz168 View Post
I'm also curious about the difference in result with your approach and mine. So just provide me with your raw data.
I will try to do it next week. (Will PM you later). Mainly I want to compare end value of equity/value at risk (x%-percentile values, value at risk for y trades) and expectancy. Currently, I'm building up my risk modelling with @risk for trading.

Great! I look forward to it.



swz168 View Post
As for @risk: the price includes a nonetime limited version. the one year maintenance means: you also get new main upgrades (not only minor bugfix for the version you bought) within one year free. This is what I remember.

A one time fee definitely makes it more palatable for retail traders.

Follow me on Twitter Started this thread Reply With Quote
 
(login for full post details)
  #88 (permalink)
 Jura   is a Vendor
 
 
Jura's Avatar
 
Posts: 774 since Apr 2010
Thanks: 2,347 given, 688 received


swz168 View Post
(..)
I'm also curious about the difference in result with your approach and mine. So just provide me with your raw data.
I will try to do it next week. (Will PM you later). Mainly I want to compare end value of equity/value at risk (x%-percentile values, value at risk for y trades) and expectancy. Currently, I'm building up my risk modelling with @risk for trading.
(...)

This will be an interesting comparison.


DarkPoolTrading View Post
Just to add my 2c to the discussion on tools and software. I think what Kevin shows is that often (usually?) process is more important than tools.


kevinkdog View Post
I agree that the process is key. If you have no process, you have no repeatability, and that makes it really hard to create consistent systems.

Can you elaborate somewhat on what you mean with 'process', as applied to trading system development, Kevin? And perhaps give an example of how you approach the process of developing a strategy? (Though there is already a lot of information about that in this thread, so I'm not sure if that question is justified)

I'm asking since I suspect that you, coming from an aerospace background, may have a whole different definition of 'process' than I, from a social science background, have.

Reply With Quote
The following user says Thank You to Jura for this post:
 
(login for full post details)
  #89 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


swz168 View Post
When I had a crash course in Monte Carlo simulation, I learned from a professor that the distribution of profit/loss is not normal distributed in the financial market. But the outcome (equity) is nearly normal distributed. So I'm wondering about your result.

Well, our comparison maybe will shed more lights into this issue.

Ahh, your professor would be very proud of you!


I should have pointed out that the chart you reference includes position sizing. That is why it has a non-normal shape.

Below is a histogram of 1 contract traded all the time. I also did 10,000 iterations to get smoother results.

As you predicted, it looks like a normal distribution.

The only difference is the spike at the lower end of the histogram. That is due to "risk of ruin" - when equity falls below a certain point, trading ceases. Without that real world restriction, the little spike would be gone.

Note to all readers: If some of this discussion is confusing or overwhelming (normal curves, random number generators, etc.), PLEASE feel free to ask questions. My philosophy is that the only dumb question is the one you are afraid to ask. So ask away. I guarantee if others have the same question(s) as you!


Follow me on Twitter Started this thread Reply With Quote
The following 4 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #90 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Jura View Post
This will be an interesting comparison.



Can you elaborate somewhat on what you mean with 'process', as applied to trading system development, Kevin? And perhaps give an example of how you approach the process of developing a strategy? (Though there is already a lot of information about that in this thread, so I'm not sure if that question is justified)

I'm asking since I suspect that you, coming from an aerospace background, may have a whole different definition of 'process' than I, from a social science background, have.


Yes, I have a defined process for developing a trading system. The basic flow chart is shown below. If you search futures.io (formerly BMT) archives, you'll find a couple of webinars I did on the topic. It is a big topic, so much so that I could actually write a book on the subject.


Follow me on Twitter Started this thread Reply With Quote
The following 3 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #91 (permalink)
 swz168 
Nuremberg, Germany
 
Experience: None
Platform: MultiCharts
Trading: Fx
 
Posts: 49 since Jun 2010
Thanks: 32 given, 59 received


kevinkdog View Post
I should have pointed out that the chart you reference includes position sizing. That is why it has a non-normal shape.

Puuh, I'm relieved. The world is normal again

In Backtest/Optimization or Simulation, in generally I don't use position sizing. Just fixed contracts. Otherwise, some metrics like "minimum capital required" is useless. Another reason with position sizing is, that good results will amplified to, let's say "perfect" result, and bad gets even worse.

Position sizing comes into play, if everything is done and when it is already decided that a strategy will be used for real trading. Then I play around with position sizing.

What is your approach? Do you apply position sizing from the beginning? Some details would be great.

Reply With Quote
The following 2 users say Thank You to swz168 for this post:
 
(login for full post details)
  #92 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


swz168 View Post
Puuh, I'm relieved. The world is normal again

In Backtest/Optimization or Simulation, in generally I don't use position sizing. Just fixed contracts. Otherwise, some metrics like "minimum capital required" is useless. Another reason with position sizing is, that good results will amplified to, let's say "perfect" result, and bad gets even worse.

Position sizing comes into play, if everything is done and when it is already decided that a strategy will be used for real trading. Then I play around with position sizing.

What is your approach? Do you apply position sizing from the beginning? Some details would be great.


I always test my systems with 1 contract, or 2 if I have some special dual exit type setup.

I see people post great equity curves, that obviously include position sizing. That is very misleading in my opinion, since if the order of the trades was a little different, the equity curve could be totally different. Also, it is hard to distinguish what is from the strategy itself (which is most important) from the position sizing (which can easily be done after the fact).

That being said, I know of one popular and successful trader (you'd probably recognize his name, and years ago he wrote a book on money management), who almost always tests with position sizing.

His theory is that if you ignore position sizing during development, you might end up with a strategy that looks good with one contract, but does not lend itself to easy position sizing. He makes a good point.

I actually have a strategy like that, very profitable, Tharp expectancy of 0.62, SQN of 3, average trade value over $2K per contract, but it is extremely difficult to position size with it unless you have a huge account. So, for 4 or 5 years I have traded it with small size. I probably would have been better off trading a worse strategy that worked better with position sizing.

So, once again, I can't say there is only one "right" way to do things.

Follow me on Twitter Started this thread Reply With Quote
The following 3 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #93 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

Week #2 is in the books. Only 2 trades taken all week, both winners.

Close to breakeven now. This reminds me of the Combine - flat to down performance.

As you can see, these strategies can be very boring.

Comparing "perfect" performance with "actual" performance, it looks like I am doing better with my actual account. Fills have been better than anticipated.









Follow me on Twitter Started this thread Reply With Quote
The following 2 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #94 (permalink)
 Jura   is a Vendor
 
 
Jura's Avatar
 
Posts: 774 since Apr 2010
Thanks: 2,347 given, 688 received


kevinkdog View Post

Thanks, that is insightful. How much does experience influence this process? Do you still need to research 100-200 ideas to come up with a good system, or does that number go down when you get better in your craft?

Edit: How long do you think that your strategies, on average, remain active? I can imagine that, if so much ideas need to be researched, the ideas that are found need to be 'active' for a longer time.


kevinkdog View Post
(..)
I actually have a strategy like that, very profitable, Tharp expectancy of 0.62, SQN of 3, average trade value over $2K per contract, but it is extremely difficult to position size with it unless you have a huge account. So, for 4 or 5 years I have traded it with small size. I probably would have been better off trading a worse strategy that worked better with position sizing.

So, once again, I can't say there is only one "right" way to do things.

If that strategy is for forex, why not use spot forex instruments instead of the much larger futures? That way, you have much more granularity over position size.

Reply With Quote
The following user says Thank You to Jura for this post:
 
(login for full post details)
  #95 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


Jura View Post
Thanks, that is insightful. How much does experience influence this process? Do you still need to research 100-200 ideas to come up with a good system, or does that number go down when you get better in your craft?

Edit: How long do you think that your strategies, on average, remain active? I can imagine that, if so much ideas need to be researched, the ideas that are found need to be 'active' for a longer time.



If that strategy is for forex, why not use spot forex instruments instead of the much larger futures? That way, you have much more granularity over position size.


As time goes on, the process does get easier with experience. It becomes much easier to eliminate bad ideas at an early phase. Plus, ideas at the start become better, since a more experienced person will know what generally works or not.

Some strategies I have developed last years. an example is below. It has traded decently for 3.5 years now. But this past year has been flat to down. Others I have traded for 5+ years. One I've been trading a year, but the last 6 months have been awful. I guess I am saying there is no predictable "shelf life" for these strategies. The key is to have new ones ready to take the place of any strats that turn bad.






Regarding forex: The strategy I am referring to unfortunately does not trade forex. It is an ag seasonal program, trades only 16 times per year, been going great since I took it live, but like I said does not lend itself to position sizing. Its equity curve is below.


Follow me on Twitter Started this thread Reply With Quote
The following 5 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #96 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received

Sometimes, after 20 plus years of trading, so tasks become so rote and routine to me that I forget that I had to learn them at one point. Such is the case with creating equity and drawdown curves. It is simple to me, but to someone who has never done it before, it can be a daunting task.

So, in this post I'll go thru the math behind creating an equity curve and a drawdown curve. If you use Excel, you can download the spreadhseet at the end to help you out.

Equity Curve

The equity curve can be built on a closed trade by trade basis, or on any timescale you wish. I like using a daily equity chart, in part to eliminate the noise from intraday price changes. You can use your daily account statement to get your current equity balance.

Here is exactly how to build an equity cruve based on daily data:

Day 0 Equity = Initial Starting Balance
Day 1 Equity = Day 0 Equity + Change in Equity During Day 1
Day 2 Equity = Day 1 Equity + Change in Equity During Day 2

...And So On...

Then, you simply plot the Day X Equity values, and you have your equity curve.



Drawdown Curve

The drawdown curve is the difference, on any given day, between that day's equity, and the maximum equity up to that point. So, let's say an account starts out with $10,000 on Day 0. On day 1, it hits a new equity high of $10,500. The drawdown on Day 1, since it is a new equity high, is $0. On Day 2, let's say the equity falls to $9,700. Now, the drawdown on Day 2 is $10,500 - $9,700 = $800. And so it goes through the rest of the days. On days where a new equity is reached, the drawdown will simply be $0. On all other days, the drawdown will be the difference between that day's equity, and the maximum equity up until that point.

Day X Drawdown = Minimum Of $0 or (Day X equity - Max Equity From Day 0 to Day X)



Excel Spreadsheet

I created a simple Excel spreadsheet, with chart, that does all these calculations for you. You just have to enter your starting balance, and then the daily equity changes. If you know Excel, you can easily add to this and make the spreadsheet an even better tool.


I hope this helps some of you!

Attached Files
Register to download File Type: xls sample curves.xls (135.5 KB, 61 views)
Follow me on Twitter Started this thread Reply With Quote
The following 5 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #97 (permalink)
 lifeguardsteve88 
Pompano Beach, Florida, USA
 
Experience: Beginner
Platform: S5 Trader, Ninja Trader
Broker: Stage5/ IQfeed
Trading: M6E, M6A, 6E, 6A, ES
 
lifeguardsteve88's Avatar
 
Posts: 159 since Jun 2013
Thanks: 272 given, 115 received


kevinkdog View Post
Sometimes, after 20 plus years of trading, so tasks become so rote and routine to me that I forget that I had to learn them at one point.

I think that sometimes when people say that all you need is a simple bar chart to trade (and perhaps one or two indicators that work with your particular style of trading), they are forgetting that to get to that point, one must gain a whole bunch of knowledge (over time) as to what works and what doesn't. And also spend a whole lot of screen time to be able to put to practice all that knowledge. I'm barely a couple of months into all this and there are so many things that now just seem like "common knowledge" or "common sense", that two months ago were giving me a major head ache just trying to grasp. I think that once you do have that knowledge base, then it is simple, in once sense.... but then there is that next step of mastering your on psychology to be able to put that knowledge to work. I'm a lot further along than day one, but see that the road ahead as a lot longer than I was imagining.

Thanks for this post, and all the other great information you pass along for all of us.

Cheers!

Visit my futures io Trade Journal Reply With Quote
The following user says Thank You to lifeguardsteve88 for this post:
 
(login for full post details)
  #98 (permalink)
 kevinkdog   is a Vendor
 
 
Posts: 3,003 since Jul 2012
Thanks: 1,598 given, 5,965 received


lifeguardsteve88 View Post
I think that sometimes when people say that all you need is a simple bar chart to trade (and perhaps one or two indicators that work with your particular style of trading), they are forgetting that to get to that point, one must gain a whole bunch of knowledge (over time) as to what works and what doesn't. And also spend a whole lot of screen time to be able to put to practice all that knowledge. I'm barely a couple of months into all this and there are so many things that now just seem like "common knowledge" or "common sense", that two months ago were giving me a major head ache just trying to grasp. I think that once you do have that knowledge base, then it is simple, in once sense.... but then there is that next step of mastering your on psychology to be able to put that knowledge to work. I'm a lot further along than day one, but see that the road ahead as a lot longer than I was imagining.

Thanks for this post, and all the other great information you pass along for all of us.

Cheers!


You make some good points, especially for a new trader.


A lot of people think that trading is getting to a point and then staying there. As if being a good trader is a destination, meaning - once you are at that point, you have "arrived" and can just take it easy.

I've been at that point many times - where I thought "wow I've made it as a trader." Coincidentally (or not), soon afterwards I almost always found myself licking my market inflicted wounds.

Now I keep in mind that there is no destination, no arriving to a land of sunshine and never ending profits. The journey itself is the destination, and it is a never ending struggle to stay sharp and ahead of the pack.

Follow me on Twitter Started this thread Reply With Quote
The following 2 users say Thank You to kevinkdog for this post:
 
(login for full post details)
  #99 (permalink)
 swz168 
Nuremberg, Germany
 
Experience: None
Platform: MultiCharts
Trading: Fx
 
Posts: 49 since Jun 2010
Thanks: 32 given, 59 received

I've done a Monte Carlo Simulation with the Data Kevin provided me.

I'm currently building up my own Monte-Carlo-Risk model for trading. Currently, I can only provide the metrics below. As already mentioned , I use the Excel add-on @risk.

The first two simulation have following inputs:

Data: 614 trades
Distribution: Discrete! (I will do a simulation with fitted data distribution another time)
Iterations: 10.000
Initial Capital: 10.000




Let's have a look at the Standard Risk scenario (Sim 1): It tells: The probability that the Equity will not fall below 10.990 is 5% (after 100 trades done). Or: the probability that the equity will exceed 10.990 is 95%.

The last two lines shows the unexpected risk. 95% is the standard scenario risk. And 99% the stress scenario risk.
I have named it Value at Risk "Drawdown", because I compare it to real drawdown. This figure gives a orientation, when to stop a strategy in real trading (depending on your own risk appetite)

Notice: The calculation is totally different to real draw down calculation! Value at Risk "Drawdown" has a weakness, the the more trades, the smaller the VaR-DD (if strategy shows a positive expectancy). So I have to compare simulations with smaller amount of trades with simulation with higher amount of trades (sim 1 and sim 2).

Simulations with smaller number of trades is absolutely necessary. It gives you more detail about the risk. Because if you have a positive expectancy, naturally the end values of an Monte Carlo simulation will be better with higher number of trades.

Interpretation Example for VaR-DD: The probabilty that the drawdown is 39% or larger is 5%.

The most important figure for me is the VaR-DD. All other metrics is more or less playing with numbers.
@Kevin: I've first done a discrete distribution simulation, so that we can directly compare our figures. Could you do the same simulation with your method with the above inputs, so we can compare?

Reply With Quote
The following 3 users say Thank You to swz168 for this post:
 
(login for full post details)
  #100 (permalink)
 swz168 
Nuremberg, Germany
 
Experience: None
Platform: MultiCharts
Trading: Fx
 
Posts: 49 since Jun 2010
Thanks: 32 given, 59 received

If you search for an alternative for @risk:

I can recommend SimulAr. I think it is inspired by @risk because the menu is designed very similar. It covers all basic functions and it is free (read the licences agreement!).

It is powerful Monte-Carlo Simulation add-on for Excel. If you think time is money, then @risk is the right tool (I think Oracle has also a risk tool, but I have never tested it).

Here is the website: SimulAr: Montecarlo simulation in Excel

Reply With Quote
The following 4 users say Thank You to swz168 for this post:


futures io Trading Community Trading Journals > Taking a Trading System Live


Last Updated on January 6, 2016


Upcoming Webinars and Events
 

NinjaTrader Indicator Challenge!

Ongoing

The Cold Hard Truth: Maybe I Am Not Good Enough w/Chris Gray @ Earn2Trade

Elite only

HIRO Indicator by Spotgamma in Bookmap w/Brent Kochuba @ SpotGamma

Elite only

NEW BlackBird Features + FOREX Support w/Jeremy Tang @ SharkIndicators

Elite only
     



Copyright © 2021 by futures io, s.a., Av Ricardo J. Alfaro, Century Tower, Panama, Ph: +507 833-9432 (Panama and Intl), +1 888-312-3001 (USA and Canada), info@futures.io
All information is for educational use only and is not investment advice.
There is a substantial risk of loss in trading commodity futures, stocks, options and foreign exchange products. Past performance is not indicative of future results.
no new posts