Auto trading is only thing that conquered my darkside - futures io
futures io



Auto trading is only thing that conquered my darkside


Discussion in Psychology and Money Management

Updated
      Top Posters
    1. looks_one liquidcci with 71 posts (65 thanks)
    2. looks_two monpere with 54 posts (53 thanks)
    3. looks_3 sidney7g with 45 posts (1 thanks)
    4. looks_4 Xeno with 33 posts (11 thanks)
      Best Posters
    1. looks_one RM99 with 1.1 thanks per post
    2. looks_two monpere with 1 thanks per post
    3. looks_3 liquidcci with 0.9 thanks per post
    4. looks_4 Xeno with 0.3 thanks per post
    1. trending_up 52,332 views
    2. thumb_up 208 thanks given
    3. group 34 followers
    1. forum 358 posts
    2. attach_file 16 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
 

Auto trading is only thing that conquered my darkside

(login for full post details)
  #301 (permalink)
 liquidcci 
Austin, TX
 
Experience: Master
Platform: ninjatrader, r-trader
Trading: NQ, CL
 
liquidcci's Avatar
 
Posts: 865 since Jun 2011
Thanks: 610 given, 1,070 received


andyb1979 View Post
Hey liquidcci - just to say I agree totally with your position. I'm a software engineer with phd in signal processing and as my bias gets in the way of trading successfully, I'm moving over to full-auto.

I'd appreciate any advice while doing this ; )

I have at present 3 algos that give good returns on the S&P500. Stock indices I know a bit about. However I'd like to move to Forex. FX has a totally different personality and the strategies I use on stock indices plain don't work on FX.

I'd be more than happy to trade a bit of information / collaborate etc on developing winning algos : )

andy welcome to the AT club. As a software engineer you should be well suited to it as a finite set of rules will make sense to you.

As far as FX goes I don't trade FX so I am not qualified to answer. I primarily trade CL Crude which you should look at.

"The day I became a winning trader was the day it became boring. Daily losses no longer bother me and daily wins no longer excited me. Took years of pain and busting a few accounts before finally got my mind right. I survived the darkness within and now just chillax and let my black box do the work."
Started this thread Reply With Quote
 
(login for full post details)
  #302 (permalink)
 monpere 
Bala, PA, USA
 
Experience: Intermediate
Platform: NinjaTrader
Broker: Mirus, IB
Trading: SPY, Oil, Euro
 
monpere's Avatar
 
Posts: 1,858 since Jul 2010
Thanks: 300 given, 3,339 received


andyb1979 View Post
Hey liquidcci - just to say I agree totally with your position. I'm a software engineer with phd in signal processing and as my bias gets in the way of trading successfully, I'm moving over to full-auto.

I'd appreciate any advice while doing this ; )

I have at present 3 algos that give good returns on the S&P500. Stock indices I know a bit about. However I'd like to move to Forex. FX has a totally different personality and the strategies I use on stock indices plain don't work on FX.

I'd be more than happy to trade a bit of information / collaborate etc on developing winning algos : )

If you have a strategy that works well on an instrument, there should be no reason (other then account requirements) to move it to another market, that would be a mistake. In doing so, you have to understand that, outside of the code itself, you are starting the proving process all over again from scratch. Every market trades differently, all the work you have done in getting your strategy to work on one instrument/market will not guarantee the same results on any other.

It's perfectly fine to experiment with the strategy with new instruments etc, but if you want to be profitable, then keep you profitable strategies, and work on scaling them with the current instrument, while you experiment with other markets.

Reply With Quote
The following 2 users say Thank You to monpere for this post:
 
(login for full post details)
  #303 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Good idea. I used to trade Crude spot (discretionary) and had some success. I found it cycled/trended nicely and was fairly predictable. One thing about FX is its a totally different beast to say stock indices. It's noisy - breakouts come out of no-where then reverse?!

Reading the pages, I got to ~10 then gave up. Save to say agree 100% with discretionary traders often being dead men walking and that its a numbers/probability game after all. A rudimentary understanding of probabillity can really help when understanding why your system has an edge that you can't see when you obey subjective trading rules and why it behaves differently in live than backtest mode (e.g. run monte carlo over its profit/loss probabililty and win:loss ratio to generate 100s of random equity curves that are all different yet use the same inputs).

Edit: yes monpere agree why move? But then I'm at the beginning of this research so now is the tmie to play around. Besides trading is an arms race, once I have one algo on one asset class that I am willing to run live I will need to have another in the pipeline close behind it. Also understanding other asset classes is a key to diversification.

Monpere I notice you mentioned in posts before "if I could program what my eye sees". How have you progressed on this front? You know rules-based algos (ie: using if-else) will likely not help you but there are an array of non-linear software techniques that can (support vector machines, AI, fuzzy logic etc)

Reply With Quote
The following user says Thank You to andyb1979 for this post:
 
(login for full post details)
  #304 (permalink)
 furytrader 
Chicago, IL USA
 
Experience: Intermediate
Platform: MultiCharts + CTS T4
Broker: Advantage Futures, IQFeed.net
Trading: YM, ES, EU, US, S
 
furytrader's Avatar
 
Posts: 153 since Jun 2011
Thanks: 109 given, 146 received

I know I'm late to the discussion here but ...

Although I don't "auto-trade" (in the sense that the computer actually enters the orders for me), I do trade using strict trading rules that are based on sound market principles and have been back-tested. Previously, I traded using a pure discretionary approach and was a middling trader at best. I found that I couldn't hold onto winners very effectively when I traded on a discretionary basis; going to model-based trading, my risk tolerance and my ability to hold onto winners has increased tremendously.

My trading has definitely improved as a result.

Reply With Quote
The following user says Thank You to furytrader for this post:
 
(login for full post details)
  #305 (permalink)
 liquidcci 
Austin, TX
 
Experience: Master
Platform: ninjatrader, r-trader
Trading: NQ, CL
 
liquidcci's Avatar
 
Posts: 865 since Jun 2011
Thanks: 610 given, 1,070 received


furytrader View Post
I know I'm late to the discussion here but ...

Although I don't "auto-trade" (in the sense that the computer actually enters the orders for me), I do trade using strict trading rules that are based on sound market principles and have been back-tested. Previously, I traded using a pure discretionary approach and was a middling trader at best. I found that I couldn't hold onto winners very effectively when I traded on a discretionary basis; going to model-based trading, my risk tolerance and my ability to hold onto winners has increased tremendously.

My trading has definitely improved as a result.

furytrader the key really is the approach you have mentioned. At some point you might look at getting your system coded as that can provide another level of benefits to a systems based trader.

"The day I became a winning trader was the day it became boring. Daily losses no longer bother me and daily wins no longer excited me. Took years of pain and busting a few accounts before finally got my mind right. I survived the darkness within and now just chillax and let my black box do the work."
Started this thread Reply With Quote
 
(login for full post details)
  #306 (permalink)
 liquidcci 
Austin, TX
 
Experience: Master
Platform: ninjatrader, r-trader
Trading: NQ, CL
 
liquidcci's Avatar
 
Posts: 865 since Jun 2011
Thanks: 610 given, 1,070 received


monpere View Post
If you have a strategy that works well on an instrument, there should be no reason (other then account requirements) to move it to another market, that would be a mistake. In doing so, you have to understand that, outside of the code itself, you are starting the proving process all over again from scratch. Every market trades differently, all the work you have done in getting your strategy to work on one instrument/market will not guarantee the same results on any other.

It's perfectly fine to experiment with the strategy with new instruments etc, but if you want to be profitable, then keep you profitable strategies, and work on scaling them with the current instrument, while you experiment with other markets.

So true. When going from market to market at minimum I usually have to adjust my stops and targets. But often I have to adjust many other parameters to make it work.

"The day I became a winning trader was the day it became boring. Daily losses no longer bother me and daily wins no longer excited me. Took years of pain and busting a few accounts before finally got my mind right. I survived the darkness within and now just chillax and let my black box do the work."
Started this thread Reply With Quote
 
(login for full post details)
  #307 (permalink)
 furytrader 
Chicago, IL USA
 
Experience: Intermediate
Platform: MultiCharts + CTS T4
Broker: Advantage Futures, IQFeed.net
Trading: YM, ES, EU, US, S
 
furytrader's Avatar
 
Posts: 153 since Jun 2011
Thanks: 109 given, 146 received

Thanks for the feedback - actually, my system is coded and it generates specific signals on the screen - however, I prefer to enter the order myself. I know, I don't have the courage yet to make the jump to straight auto-trading!

Reply With Quote
 
(login for full post details)
  #308 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Semi-auto then. Yes another very valid method. I have a colleague who trades privately with a semi-auto system. He devised the rules himself and obeys them strictly. He tells me changing from discretionary to system based (even though he pulls the trigger) made the difference between unprofitable and profitable

Reply With Quote
 
(login for full post details)
  #309 (permalink)
 eudamonia 
Sacramento, CA
 
Experience: None
Platform: None
Broker: ADM and Sierra Charts
Trading: ES, CL
 
eudamonia's Avatar
 
Posts: 315 since Jul 2010
Thanks: 308 given, 448 received

Trading by the seat of your pants isn't discretionary trading it's gambling. Discretionary traders have a method that is rules based and they follow their rules to execute their edge. They just happen to push the buy/sell buttons themselves.

Having started off my trading career with the intent to trade fully AT and having moved to trading discretionarily, I can now say that some things cannot adequately be automated. Things like divergences, support/resistence zones based on market profile, and the like are ultimately patterns that a computer is unlikely to be able to replicate the same as an individual. And while possible, programming a system that can factor in longer term biases while remaining adaptive is a massive undertaking.

Lastly, I think most traders vastly underestimate the brain power of their competition and the overall effort required to trade AT. I believe that it is safe to say that unless you possess a far above average understanding of mathmatics and programming (not just a couple undergraduate courses but a MS or PhD level or the equivalent in work experience) you are probably wasting your time trying to create an AT system that trades more than a couple times a month. A good test of this is whether you possess the programming/math knowledge to create your own trading platform, optimization (using some sort of hill climbing/adaptive alg.), and ATI interface with your broker. That's not to say there aren't some very good out of the box tools available but my experience has been that the successful AT traders can and often do customize anything that might affect their systems. Needless to say they fully understand every process that can and often does interact/interfere with their AT trading.

Obviously you may be the exception to the rule, but from my 4 years of experience pursuing AT I was never able to be half as good as my discretionary trading and I of course thought I was above average intelligence.

Visit my futures io Trade Journal Reply With Quote
The following user says Thank You to eudamonia for this post:
 
(login for full post details)
  #310 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received


eudamonia View Post
MS or PhD level or the equivalent in work experience ... A good test of this is whether you possess the programming/math knowledge to create your own trading platform, optimization (using some sort of hill climbing/adaptive alg.), and ATI interface with your broker.

Yes and yes, but even so its not easy by any stretch of the imagination. The part I find hard is not the technology side, its the trading side. Coming up with rules/methods for trading that are not subjective and work is actually very difficult. As an example, a statement "uptrends consist of higher highs and lows" may be subjective. In this case how would you define highs/lows? If you can quantify it then you can auto-trade it. Failing that semi auto where you devise indicators to spot the opportunity in the noise. It's likely you just need a non-linear decision maker. If/else statements won't cut it as the market is more complex than that.

Yes agree the average chap will have a hard time AT. However the average chap will also have a hard time discretionary trading - isn't it 90% lose or something?


eudamonia View Post
Obviously you may be the exception to the rule, but from my 4 years of experience pursuing AT I was never able to be half as good as my discretionary trading and I of course thought I was above average intelligence.

I personally would never be so presumptious as to think I'm the exception to the rule. Financial markets are good at humbling you if you think you're a hotshot. However, I have learned that my discretionary is rubbish so ready to try something else ;0

So far in a month of R&D, backtesting etc... I've learned loads about what is and is not statistically viable. I dare say the computer taught me more about trading in that time than I knew from years previously. Things like what is the effect on long term profits of an x% stoploss, or the effect on drawdown of taking partial profits early. Things like what indicator combinations are useful and which are useless. I've found a very simple combination of 1x trend indicator (to gauge long or short) and 1x cycle indicator (for entry) combined with stop, take profit & trailing stop to be effective. If I didn't test these assumptions I'd easily get discouraged when the first losses came.

Reply With Quote
 
(login for full post details)
  #311 (permalink)
 GT22 
Currently Inactive
 
 
Posts: 25 since Aug 2011

I am hoping that it is not necessary for me to be a rocket scientist to automate some special situation trades. Forex is a treasure trove of special situation opportunities! The problem: I am not fast enough to trade them manually mostly because it works best to trade these opportunities with multiple pairs.

At this point I have these pieces developed:

The ability to continuously scan across 20+ pairs

Various custom regressions and automatic algebraic trendlines

And most importantly several solid concepts.


I do not trust MT4 for backtesting since many of my concepts do not require stops exceeding 10 to 15 pips. I also want a portable solution so moving to an external dll seems to be required. Looks like my limited C and Pascal capabilities will need to be upgraded before I can continue.

Reply With Quote
 
(login for full post details)
  #312 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Start off with MT4 to prove the concept in a backtested EA then look to replicate with standalone code or a more solid platform.

But yes its a lot of work. Some brokers (Interactive) will provide you decent APIs (C#/Java) for order generation, instrument definition and historical / live data subscription etc...

Reply With Quote
 
(login for full post details)
  #313 (permalink)
 GT22 
Currently Inactive
 
 
Posts: 25 since Aug 2011


andyb1979 View Post
Yes and yes, but even so its not easy by any stretch of the imagination. The part I find hard is not the technology side, its the trading side. Coming up with rules/methods for trading that are not subjective and work is actually very difficult. As an example, a statement "uptrends consist of higher highs and lows" may be subjective. In this case how would you define highs/lows? If you can quantify it then you can auto-trade it. Failing that semi auto where you devise indicators to spot the opportunity in the noise. It's likely you just need a non-linear decision maker. If/else statements won't cut it as the market is more complex than that.

I have been able to define the swing highs/lows so you should be able to also. You can also do a lot of things with them that make much more sense than the traditional indicators. Other than trying to use some unique regression stuff to quantify what my eyes are seeing, I do not use any indicators.


andyb1979 View Post
So far in a month of R&D, backtesting etc... I've learned loads about what is and is not statistically viable. I dare say the computer taught me more about trading in that time than I knew from years previously. Things like what is the effect on long term profits of an x% stoploss, or the effect on drawdown of taking partial profits early. Things like what indicator combinations are useful and which are useless. I've found a very simple combination of 1x trend indicator (to gauge long or short) and 1x cycle indicator (for entry) combined with stop, take profit & trailing stop to be effective. If I didn't test these assumptions I'd easily get discouraged when the first losses came.

If you are going to pursue FX, ditching the indicators might be a good idea. For me, the indicators took my attention away from the bars on the chart where the real information is. But everyone is different so definitely pursue your own path.

Reply With Quote
 
(login for full post details)
  #314 (permalink)
 Lornz 
Oslo, Norway
 
Experience: Advanced
Platform: CQG, Excel
Trading: CL
 
Lornz's Avatar
 
Posts: 1,198 since Apr 2010


GT22 View Post
I have been able to define the swing highs/lows so you should be able to also. You can also do a lot of things with them that make much more sense than the traditional indicators. Other than trying to use some unique regression stuff to quantify what my eyes are seeing, I do not use any indicators.

Shhh!

I've always been amazed by how the obvious is so often overlooked by most... Basic concepts, using (often) simple arithmetic, can be quite useful...

Visit my futures io Trade Journal Reply With Quote
 
(login for full post details)
  #315 (permalink)
 GT22 
Currently Inactive
 
 
Posts: 25 since Aug 2011


andyb1979 View Post
Start off with MT4 to prove the concept in a backtested EA then look to replicate with standalone code or a more solid platform.

But yes its a lot of work. Some brokers (Interactive) will provide you decent APIs (C#/Java) for order generation, instrument definition and historical / live data subscription etc...

I believe that my stops will be too tight for backtesting on the 1 minute bars. Perhaps not. Perhaps it can still be used to validate the concept, then I would have to port it to test the desired stop placements.

Yes, Oanda also has an API plus I want to add at least one more broker. The goal of moving to direct APIs makes more sense to me than trading through MT4, MultiCharts, NeoTicker, NinjaTrader, etc.

Reply With Quote
 
(login for full post details)
  #316 (permalink)
 monpere 
Bala, PA, USA
 
Experience: Intermediate
Platform: NinjaTrader
Broker: Mirus, IB
Trading: SPY, Oil, Euro
 
monpere's Avatar
 
Posts: 1,858 since Jul 2010
Thanks: 300 given, 3,339 received


andyb1979 View Post
Good idea. I used to trade Crude spot (discretionary) and had some success. I found it cycled/trended nicely and was fairly predictable. One thing about FX is its a totally different beast to say stock indices. It's noisy - breakouts come out of no-where then reverse?!

Reading the pages, I got to ~10 then gave up. Save to say agree 100% with discretionary traders often being dead men walking and that its a numbers/probability game after all. A rudimentary understanding of probabillity can really help when understanding why your system has an edge that you can't see when you obey subjective trading rules and why it behaves differently in live than backtest mode (e.g. run monte carlo over its profit/loss probabililty and win:loss ratio to generate 100s of random equity curves that are all different yet use the same inputs).

Edit: yes monpere agree why move? But then I'm at the beginning of this research so now is the tmie to play around. Besides trading is an arms race, once I have one algo on one asset class that I am willing to run live I will need to have another in the pipeline close behind it. Also understanding other asset classes is a key to diversification.

Monpere I notice you mentioned in posts before "if I could program what my eye sees". How have you progressed on this front? You know rules-based algos (ie: using if-else) will likely not help you but there are an array of non-linear software techniques that can (support vector machines, AI, fuzzy logic etc)

I am trading in semi auto mode because as of yet, I still am not fully satisfied with the accuracy of the signal identification of my code, but I have made some progress lately when I have time to play with the code. The only thing that is not automated is the pulling of the trigger, because a percentage of the signals identified by code are not valid as my brain sees them. This is due to the fact that my method is pretty visual.

Reply With Quote
 
(login for full post details)
  #317 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Yeah semi-auto is great also. basically what I mean when I say I agree with the original poster is you need a system. Either you pull the trigger (and are very accurate) or the computer does, but either way, if the system is tested + obeyed its a recipe for success.

Seat of the pants in my experience, is not!! You may be different but I can't hold my nerve when discretionary trading. Unless its one of those screaming buy moments like oil at $35 : )

Ok I'm making my first foray into Ninjatrader. I've been backtesting using R then later OpenQuant (evaluation) looking for a trading platform to make that first auto trade. I'm currently using NT as I am impressed with the support/community, however I am not impressed with how buggy it is. Seriously guys, writing software that works is not that hard ...

Here's a printout of one of my strategies in the optimiser. I want to optimise this using Piersh Genetic (I have installed the extension) and maximise the SQN (System Quality Number - again extension). I am testing on NYSE:SSO daily bars (S&P500 leveraged index fund). To my dismay the optimiser outputs only one row. Even when in "Default" mode.

Whats going on here? NT can't be that crap, surely its my mistake.

Attached Thumbnails
Click image for larger version

Name:	NT only one result.png
Views:	118
Size:	176.3 KB
ID:	47501  
Reply With Quote
 
(login for full post details)
  #318 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received


Quoting 
I have been able to define the swing highs/lows so you should be able to also. You can also do a lot of things with them that make much more sense than the traditional indicators. Other than trying to use some unique regression stuff to quantify what my eyes are seeing, I do not use any indicators.

Hey there, yes I wanted to de-subjectivise this process so I used a slow Fisher Transform (John Elhers) to detect cycle tops/bottoms, then drew a Manhattan chart on the turning points. I'm in the process of researching *how* to use this information in trading systems. I'm thinking a gann based method, like buy new highs, sell new lows. Or at least use previous low (or high) as stop-loss when long (short) and accumulating positions

Your thoughts welcome

Attached Thumbnails
Click image for larger version

Name:	Fisher-based manhatten.png
Views:	118
Size:	77.2 KB
ID:	47502  
Reply With Quote
 
(login for full post details)
  #319 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received


andyb1979 View Post
Whats going on here? NT can't be that crap, surely its my mistake.

sorry, im being a total thicko. I forgot to click the tab to show the top 20 :S

Reply With Quote
 
(login for full post details)
  #320 (permalink)
 liquidcci 
Austin, TX
 
Experience: Master
Platform: ninjatrader, r-trader
Trading: NQ, CL
 
liquidcci's Avatar
 
Posts: 865 since Jun 2011
Thanks: 610 given, 1,070 received


andyb1979 View Post
sorry, im being a total thicko. I forgot to click the tab to show the top 20 :S

NT has its issues although I find NT7 to work quite well and to be stable. Trading software you just have know limitations of each vendor. All have pluses and minuses. In many ways comes down to what you need. What I do works great with NT but may not work for someone else. One thing is for sure right strategy you can make a lot of money with NT.

"The day I became a winning trader was the day it became boring. Daily losses no longer bother me and daily wins no longer excited me. Took years of pain and busting a few accounts before finally got my mind right. I survived the darkness within and now just chillax and let my black box do the work."
Started this thread Reply With Quote
 
(login for full post details)
  #321 (permalink)
 zer0 
Chicago, IL/USA
 
Experience: Advanced
Platform: CTS T4/Sierra Chart
Trading: Futures
 
zer0's Avatar
 
Posts: 139 since Jun 2011
Thanks: 49 given, 191 received

Amazing, isn't it? Anyway, good job, GT22. Now stop talking lol.


Lornz View Post
Shhh!

I've always been amazed by how the obvious is so often overlooked by most... Basic concepts, using (often) simple arithmetic, can be quite useful...


Reply With Quote
 
(login for full post details)
  #322 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

True. I'm fighting the urge to write my own basic backtester/execution engine, but that'll take me on a loooong road to no trading profits. Better to focus on the strategy.

Reply With Quote
 
(login for full post details)
  #323 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Here are a couple screenshots to illustrate the point that autotrading, for me, is the only way. The first is the result of a walk forward test of a strategy I'm working on. The strategy was trained using a stochastic optimiser over 2007-2010 15 minute bars of EURUSD. This found the best parameters for maximum PnL / drawdown (aka smoothest equity curve).

Next I re-ran the test for the full dataset, 2007-2012 with the optimum parameters. Ok the equity curve is still advancing and still smooth. All good I think.

So I zoom in to the chart to see the sort of trades its taking. WTF? Tiny scalps that make no sense. its missing huge moves and taking 20 pips here and there. No way could I personally trade this, but the software says statistically this method is extremely good. Stable for years and slowly climbing the equity curve.

I have yet to run this strategy live - I'm still testing, but I'm encouraged by a walk forward. Whats odd is when I try to add "intuitive improvements", for instance, adding a trailing stoploss, or rules to maximise the profit, it fails. I'm only using a crude stop (very wide) on this to protect against system failure so for all intents and purposes its a no-stop system.

My first and so far best trading strategy for the EUR/USD. Also works on Stock index futures. The probabilities don't lie, even though they don't make sense ...

Attached Thumbnails
Click image for larger version

Name:	EURUSD Forward Test.png
Views:	125
Size:	149.5 KB
ID:	47561   Click image for larger version

Name:	EURUSD Forward Test2.png
Views:	111
Size:	230.6 KB
ID:	47562  
Reply With Quote
 
(login for full post details)
  #324 (permalink)
 Xeno 
UK
 
Experience: Intermediate
Platform: Ninja
Broker: Mirus/Zen
Trading: Futures - bonds, currencies, index
 
Posts: 288 since Oct 2010
Thanks: 70 given, 270 received

The first pic is too small - difficult to see what's going on. It does, maybe appear though, that there are periods of quite a few months when the strategy is flat, or in slight drawdown?

Reply With Quote
 
(login for full post details)
  #325 (permalink)
 furytrader 
Chicago, IL USA
 
Experience: Intermediate
Platform: MultiCharts + CTS T4
Broker: Advantage Futures, IQFeed.net
Trading: YM, ES, EU, US, S
 
furytrader's Avatar
 
Posts: 153 since Jun 2011
Thanks: 109 given, 146 received

I agree with the following two points made previously:

1) While the "IT" or programming aspect of developing an automated model may be hard, I find that identifying a solid trading strategy that is based on sound market principles is even harder. When developing quantitative models, it's possible to generate strong backtested results, but if the core principle of the model is not sound, you're at risk for performance drift (or complete breakdown) when trading in real-time. That's why I've shied away from using neural nets, genetic algorithms, etc. although I find the ideas behind them fascinating.

2) "I've always been amazed by how the obvious is so often overlooked by most... Basic concepts, using (often) simple arithmetic, can be quite useful..." The models I use right now in the stock indexes are not rocket science, and several came from books that I adapted to the markets I trade. The fact that several of these are based on market ideas that have been around for ages gives me more confidence to trade them. I have backtested them all as well.

Reply With Quote
 
(login for full post details)
  #326 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Larger pic of the equity curve. Please note this is the same strategy but with a few tweaks so not the same curve as the other image.

Yes some periods of drawdown, some significant. The red shaded area from 2007-2010 was my training window and green was a walk forward test.

Also note this test traded 10x contracts of EURUSD $1 mini (aka $10/pip). It makes $70,000 in a few years off a $10,000 account. Yeahhh right! I'm not believing this performance, its the increase/volatility of the equity curve I am interested in. In reality I would probably trade live 1x contract off the same account size.

@Fury Developing strategies is definitely the hardest part! The software tools only really help you do this more quickly, but one (successful) trader I know developed his method manually over a period of 3 years. Now executes manually with varying but steady success

Attached Thumbnails
Click image for larger version

Name:	EURUSD Forward Test3.png
Views:	102
Size:	58.9 KB
ID:	47563  
Reply With Quote
 
(login for full post details)
  #327 (permalink)
 Xeno 
UK
 
Experience: Intermediate
Platform: Ninja
Broker: Mirus/Zen
Trading: Futures - bonds, currencies, index
 
Posts: 288 since Oct 2010
Thanks: 70 given, 270 received

Doesn't look too bad. A few questions - what is the average trade PnL, and does it include commission? Is it market, limit or stop entry? Are most of the trades during EU/US day, or night?




andyb1979 View Post
Larger pic of the equity curve. Please note this is the same strategy but with a few tweaks so not the same curve as the other image.

Yes some periods of drawdown, some significant. The red shaded area from 2007-2010 was my training window and green was a walk forward test.

Also note this test traded 10x contracts of EURUSD $1 mini (aka $10/pip). It makes $70,000 in a few years off a $10,000 account. Yeahhh right! I'm not believing this performance, its the increase/volatility of the equity curve I am interested in. In reality I would probably trade live 1x contract off the same account size.

@Fury Developing strategies is definitely the hardest part! The software tools only really help you do this more quickly, but one (successful) trader I know developed his method manually over a period of 3 years. Now executes manually with varying but steady success


Reply With Quote
 
(login for full post details)
  #328 (permalink)
 mainstream 
Chicago, IL
 
Experience: Master
Platform: Kinetick Ninja Trader <7>
Broker: Ninja & IB
Trading: YM & Equities & Options & Mutual Funds
 
Posts: 175 since Jun 2010
Thanks: 56 given, 87 received


furytrader View Post
from books that I adapted to the markets I trade. The fact that several of these are based on market ideas that have been around for ages gives me more confidence to trade them. I have backtested them all as well.

Would you share the sources?

Reply With Quote
 
(login for full post details)
  #329 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Here are (honest) answers to your questions

Include commissions - No (Openquant doesnt support this in test mode, as far as I can tell)
Slippage - No
Orders - All at market on bar close
Expectancy - with these settings about +$35 per trade.

i.e. average of all losses + all wins / total trades is $35

So I know what you're thinking. $35 - some slippage - some commissions (about $3 per side on interactive brokers FX) = potential to go negative in live trading.

Well, its a start

I'd like to see the expectancy higher. The win:loss ratio Im pleased with at its 68%. The average win however is not far off the average loss.

I'm trying to add some better profit taking like scalp 50% at a certain level then trail the rest to see how that improves expectancy

Reply With Quote
 
(login for full post details)
  #330 (permalink)
 Xeno 
UK
 
Experience: Intermediate
Platform: Ninja
Broker: Mirus/Zen
Trading: Futures - bonds, currencies, index
 
Posts: 288 since Oct 2010
Thanks: 70 given, 270 received

That's exactly what I'm thinking. To be honest, if you see a decent equity curve, you usually find there's a problem with something, hence my questions.

I don't just see a potential for losses when live. I see it guaranteed, if I read this right. I don't know exactly how your commissions work in FX, but if it was futures (which is probably similar) then $3.5 per contract average trade is completely wiped out by about $5 commission per contract. In your case it may be $6. That's before we've even started on slippage.

Seriously, don't even bother with your equity curve until your average per contract traded is way bigger (pref much more than treble) than your commission per contract.


andyb1979 View Post
Here are (honest) answers to your questions

Include commissions - No (Openquant doesnt support this in test mode, as far as I can tell)
Slippage - No
Orders - All at market on bar close
Expectancy - with these settings about +$35 per trade.

i.e. average of all losses + all wins / total trades is $35

So I know what you're thinking. $35 - some slippage - some commissions (about $3 per side on interactive brokers FX) = potential to go negative in live trading.

Well, its a start

I'd like to see the expectancy higher. The win:loss ratio Im pleased with at its 68%. The average win however is not far off the average loss.

I'm trying to add some better profit taking like scalp 50% at a certain level then trail the rest to see how that improves expectancy


Reply With Quote
 
(login for full post details)
  #331 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Good advice.

I found the options in Openquant (thanks google) to add commissions and slippage. I added $3/side for FX (Interactive brokers) and 0.0001 slippage. That's a mere pip. I'm thinking to simulate bid/ask and slippage I should be going more in the region of 5 pips.

Here's the updated curve with the same parameters. Ok great exercise - now I can work on the "real" problem not the fake problem. Lol

FYI Lightspeed broker offers extremely high speed and low commissions. I'm thinking to use them in the end not IB.

Attached Thumbnails
Click image for larger version

Name:	EURUSD Forward Test 4.png
Views:	116
Size:	56.8 KB
ID:	47575  
Reply With Quote
 
(login for full post details)
  #332 (permalink)
 Silver Dragon 
Legendary Data Wizard!!!
Cincinnati Ohio
 
Experience: Intermediate
Platform: TastyWorks
Broker: TastyWorks
Trading: FX, Stocks, Options
 
Silver Dragon's Avatar
 
Posts: 2,011 since Feb 2011
Thanks: 5,770 given, 4,875 received

You might try MB Trading for Forex. 2.95 per 100,000. Plus you get discounts when using limit orders on their ECN. My average commission when trading 10,000 lot is less than .40 cents round trip with their setup.



andyb1979 View Post
Good advice.

I found the options in Openquant (thanks google) to add commissions and slippage. I added $3/side for FX (Interactive brokers) and 0.0001 slippage. That's a mere pip. I'm thinking to simulate bid/ask and slippage I should be going more in the region of 5 pips.

Here's the updated curve with the same parameters. Ok great exercise - now I can work on the "real" problem not the fake problem. Lol

FYI Lightspeed broker offers extremely high speed and low commissions. I'm thinking to use them in the end not IB.


nosce te ipsum

You make your own opportunities in life.
Visit my futures io Trade Journal Reply With Quote
 
(login for full post details)
  #333 (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: 50,002 since Jun 2009
Thanks: 32,468 given, 98,268 received


Xeno View Post
That's exactly what I'm thinking. To be honest, if you see a decent equity curve, you usually find there's a problem with something, hence my questions.

First thing I thought of which doesn't seem to have been brought up yet, was the "test" time frame. If the in-sample data is 2007-2010, and the out of sample data is 2010-2011, my question is - have you made any changes to this strategy since testing it on 2010-2011 data? Because once you do, then it is no longer out of sample data, but in sample data...

For example, first time you run the backtest, 2010-2011 looks bad. Hours later you emerge with tweaks, and now 2010-2011 looks good. That is the same as testing 2007-2011 all at once, and highly curve fitted.

As others have mentioned, you need to account for slippage (a market order is 'slippage' right there) and commissions. If you are taking any trades around news events, that could be a major problem too.

Last, I noticed you are using 15m bars. I don't know what you are using to backtest, I am not familiar with it. But most likely, if like some other platforms, then testing on such a big bar for such small profit targets will result in very unreliable results, due to the lack of OHLC ordering (did the H come first, or the L). If your software supports a per-trade report, look to see how many trades reach their profit target on 1 bar. Now go back and count all those as losers, not winners, and see how it looks.

Naturally you can learn these things from trading live as well. That's how I did.

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
The following 2 users say Thank You to Big Mike for this post:
 
(login for full post details)
  #334 (permalink)
 RM99 
Austin, TX
 
Experience: Advanced
Platform: TradeStation
Trading: Futures
 
Posts: 882 since Mar 2011
Thanks: 124 given, 701 received

I'll try to be diplomatic.

1) I use a $25/contract/roundtrip commission and slippage as a baseline and I don't even look twice at a strategy that doesn't yield at least 5 ticks/trade above and beyond that. Low net profit/trade systems tend to be all hat and no cattle. They get absolutely monkey stomped when going live.

2) A strategy that yields $70k over 4 years COULD be worth the risk if it truly yielded that much. My experience is that if a strategy (when backtested) yields $X, you can cut that in half to get an idea of how much it will actually yield live. That's because backtesting cannot escape at least some degree of curve fitting/opmiziation. You should do a serious robustness evaluation to determine just how optimized your system is. If you alter the setting even slightly and the strategy loses SIGNIFICANT profits, then be wary. If you alter the settings and the strategy goes negative (unless you absolutely, intentionally try to sabotage it) then run the other way.

You should always compare the anticipated return with your buy and hold alternative. What was the risk free rate of return over the last 4 years? Is the risk of trading a system worth it compared to what you could have made if you simply invested in the market at large? I've told some of my friends that tried to dabble in trading, that sometimes you'd be better off just spending the money, that way at least you know you'll get some satisfaction out of it.

3) As Mike informed you, any strategy that can complete an entire (entry/exit) cycle within a single bar increment on your chart is inherently biased.

In essence....keep working. You'll get there, but if you ever get that feeling you've developed the printing press, the Golden Goose or the money tree, come back to futures.io (formerly BMT) for some grounding

"A dumb man never learns. A smart man learns from his own failure and success. But a wise man learns from the failure and success of others."
Reply With Quote
The following 2 users say Thank You to RM99 for this post:
 
(login for full post details)
  #335 (permalink)
 Xeno 
UK
 
Experience: Intermediate
Platform: Ninja
Broker: Mirus/Zen
Trading: Futures - bonds, currencies, index
 
Posts: 288 since Oct 2010
Thanks: 70 given, 270 received

Lot of good advice coming in here from people who have been there.

In my backtesting career, I have had about four occassions where a very good equity curve and stats jumped out at me early in the testing process. On 100% of those occassions there was an error in the strategy or backtesting process. By the fourth time it happened, there was no excitement. I knew immediately what the likely explanation was. I'm afraid that's the sort of sceptical mindset you have to have.

Your good looking equity curve was a very good example of how one of the stats can look great but it's pointless getting excited until you've ticked the boxes that need to be ticked.

In anticipation of someone asking what those ticks are, here are mine

1. Enough trades for significant stats
2. Distribution of optimised parameters is robust (This is done before the test really. This is what RM99 meant when saying you should be able to tweak parameters without profit plunging)
3. avge per trade is way more than commission + typical slippage
4. Equity curve is regular
5. Drawdown acceptable in amount and duration for account size.
6. Sanity check on trades, esp big winners and losers and ones that last one bar
7. Trade data is out of sample
8. Testing period includes different market types appropriate for strategy (e.g. period is not just a bull market) - another way of looking at that is that there is a reasonably even split between long and short profits

There are others, which I go into in more detail afterwards, but that's my starting eight. It's tempting to get excited if a few of those look good, but I could develop a strategy in 15 mins that had great figures for a few of them.

Reply With Quote
The following 3 users say Thank You to Xeno for this post:
 
(login for full post details)
  #336 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Hi guys,

Wow, sage advice, Yes I get the impression I just learned a potentially hard lesson, the easy way. I also think that since this is a non-linear problem perhaps I need to up the ante slightly and attack it in a non-linear way. So far I'm using indicator combinations and a little money management and I think it shows that it may work for some time, but not all the time. Or it may work in a canned market but in the real world? ...

re some points raised, some comments;



Quoting 
For example, first time you run the backtest, 2010-2011 looks bad. Hours later you emerge with tweaks, and now 2010-2011 looks good. That is the same as testing 2007-2011 all at once, and highly curve fitted.

Yes - in the cases I was considering out of sample always looked good (ie: positive) before further optimisation, however adding commissions and now 5 pips slippage (realistic on EURUSD) my portfolio takes a bath. Back to the drawing board. Its something to be aware of though.

Just out of interest, how do you backtest properly consider once you've looked at out of sample data once, then you risk curve fitting if you tweak? I was planning to paper trade a full 6 months on an algorithm that retrained regularly, say every week.


Quoting 
You should always compare the anticipated return with your buy and hold alternative. What was the risk free rate of return over the last 4 years? Is the risk of trading a system worth it compared to what you could have made if you simply invested in the market at large? I've told some of my friends that tried to dabble in trading, that sometimes you'd be better off just spending the money, that way at least you know you'll get some satisfaction out of it.

This is a good point and precisely what financial institutions do when they estimate the volatility adjusted returns (Sharpe ratio) using the Risk Free rate (typically 10 year Gilts or Treasuries) as a basis. Also I was considering trying an alternative which is random buys/sells ie: a random walk trader. If my system can't beat random them its rubbish.

Regarding "risk free" investment, currently 100% of my money is invested in a Nationwide super saver account paying 3%. LOL. There it will stay until I am confident I have a system or preferably systems that are worth taking the risk over.


Quoting 
1. Enough trades for significant stats
2. Distribution of optimised parameters is robust (This is done before the test really. This is what RM99 meant when saying you should be able to tweak parameters without profit plunging)
3. avge per trade is way more than commission + typical slippage
4. Equity curve is regular
5. Drawdown acceptable in amount and duration for account size.
6. Sanity check on trades, esp big winners and losers and ones that last one bar
7. Trade data is out of sample
8. Testing period includes different market types appropriate for strategy (e.g. period is not just a bull market) - another way of looking at that is that there is a reasonably even split between long and short profits

Excellent list thanks. Re-runing the optimiser with commissoins and slippage I found the optimiser cleverly optimised out all the trades (aka the losses being commissions) leaving me with an optimum condition of "No trades". Very good

So, Openquant doesnt let you write your own fitness function but there is a workaround. I'm planning to use System Quality Number(SQN) which includes some weighting given to trading opportunities.

Ave per trade > commission + slippage definitely. I need to reduce trading and increase the profit target.

4, 5, 6, 7, 8 . Yes good ideas

-----

So, I feel a little dissapointed, out of my 3 strategies I'd developed in R 2 now lose my shirt and the third scrapes a living off the EURO or GBP but not worth the risk IMO.

Backing out to a daily chart its a different story and my systems are still profitable but not as good as they were (certainly more volatile). I think maybe it would be wise to drop back to the daily scale where theres more signal less noise as I get to grips with these concepts. There are plenty of opportunities to be made on Stock indices, Fx pairs and major commodities just with simple rules based strategies. A system that trades once a month may be boring yes, but we're not doing this for excitement right?

For shorter timescales I need a non-linear system. I have loads of ideas just when to implement them??!

Thanks again guys and keep up the feedback, I'm listening

Reply With Quote
 
(login for full post details)
  #337 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Ok, I took a step back and applied a trend following strategy I've developed on the daily EURUSD only. Forget 15 mins for now, its too noisy. I need to go back to basics and back to the daily chart.

So this strategy is simple, it buys dips in an uptrend using SMA as a trend indicator (Gradient of SMA = trend) and a fast oscillator for entry. Now the secret sauce was this. I added in a take-profit target to close 50% of the position and a stoploss (initially placed below the recent low, moved to breakeven after take-profit was hit). I ran this through the optimiser. Now I didn't optimise ANY parameters of indicators - just the profit target. Once the profit target is hit, it then runs a trailing stop on the remaining 50%.

The result - profitable! This is with commissions and 5 pips slippage. I changed all orders to limit orders (stop, entry, take profit, trailing). Stops are updated daily (on bar close).

Now this is the best bit. I ran the same strategy over a basket of stocks without any re-optimisation and its profitable on every single one. Ok some more volatile returns than others but not bad at all. Now I'm thinking its too good to be true (heh)

Next up.

- My trailing stop implementation is a simple %. I want to use an ATR stop (e.g. the SuperTrend) to maximise profit
- More testing. Try to break it bad - higher slippage, higher commissions
- More sanity checking. Simulate on wider basket of stocks/instruments
- Paper trading on live

I'll keep you updated

Attached Thumbnails
Click image for larger version

Name:	trendfollowing stats.png
Views:	111
Size:	105.6 KB
ID:	47673  
Reply With Quote
The following user says Thank You to andyb1979 for this post:
 
(login for full post details)
  #338 (permalink)
 RM99 
Austin, TX
 
Experience: Advanced
Platform: TradeStation
Trading: Futures
 
Posts: 882 since Mar 2011
Thanks: 124 given, 701 received


andyb1979 View Post
Ok, I took a step back and applied a trend following strategy I've developed on the daily EURUSD only. Forget 15 mins for now, its too noisy. I need to go back to basics and back to the daily chart.

So this strategy is simple, it buys dips in an uptrend using SMA as a trend indicator (Gradient of SMA = trend) and a fast oscillator for entry. Now the secret sauce was this. I added in a take-profit target to close 50% of the position and a stoploss (initially placed below the recent low, moved to breakeven after take-profit was hit). I ran this through the optimiser. Now I didn't optimise ANY parameters of indicators - just the profit target. Once the profit target is hit, it then runs a trailing stop on the remaining 50%.

The result - profitable! This is with commissions and 5 pips slippage. I changed all orders to limit orders (stop, entry, take profit, trailing). Stops are updated daily (on bar close).

Now this is the best bit. I ran the same strategy over a basket of stocks without any re-optimisation and its profitable on every single one. Ok some more volatile returns than others but not bad at all. Now I'm thinking its too good to be true (heh)

Next up.

- My trailing stop implementation is a simple %. I want to use an ATR stop (e.g. the SuperTrend) to maximise profit
- More testing. Try to break it bad - higher slippage, higher commissions
- More sanity checking. Simulate on wider basket of stocks/instruments
- Paper trading on live

I'll keep you updated

You are getting it.

You can optimize profit taking and exit management with much less curve fitting effect than if you try to optimize the entries.

There are thousands of tricks/methods for optimizing and managing the trade exit.

Breakeven points. Scale out points. Scale in points. Simple trailing stops. Custom trailing stops (that only activate at certain levels and adjust, like parabolics and shrinking stops).

By trying combinations of exits, you'll find that systems that were marginal can become viable.

When I optimize I tend to leave the entry signals alone. I don't touch them. I simply try to see how much money I can squeeze out of the entries I do have.

This way, there's still curve fitting involved...but usually, your live trading extracts some fraction of the optimized result, but it's usually a very good chance that it will be profitable. (i.e. the optimized results might have been $10k, and in reality you were able to only capture $7k). The live results will almost always be some number less than the "optimal solution."

This is another way of saying "let your winners run and cut your losers short." A winning trade will be a winning trade, it's just a matter of how much.

Also, be very wary of optimizing your stop loss amounts. You can actually curve fit to the point that a slight adjustment means huge swings in profitability, so do a robustness check and leave yourself room to breathe. Just because your optimization says that the optimal stop amount is 31 ticks, doesn't mean that's the best choice.

What happens is that the optimizer recognizes that there was a single trade that drew down 30 ticks and then recovered for a nice large (sometimes out of tolerance) winner. In essence, had your stop been 30 ticks, it would have been a huge swing, cause it would have been a loser instead of a winner (and a big winner at that).

In the end, I try to leave stops and entry signals fairly static and use a combination of profit exits and trails to wring out as much profit as a system will allow.

The next step is discovering that all systems and their efficiency are dependent upon the entry. But I find it's much easier and expeditious to start by optimizing and focusing on exits....when you REALLY get good is when you can find a significant and consistent edge

"A dumb man never learns. A smart man learns from his own failure and success. But a wise man learns from the failure and success of others."
Reply With Quote
The following 2 users say Thank You to RM99 for this post:
 
(login for full post details)
  #339 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Thanks!

Ok good so its not impossible then (to create a working trading strategy). Yes from speaking to several traders I have been told to take a percentage of profit early and move a stop to break even. Only when I implemented this in software and simulated it did I see how powerful it actually is.

re: Optimisation and curve fitting, I get it. I feel more confident that I did not touch the entry rules, just exit, that this has a chance of working live. Also I plan when I optimise to output the optimisation parameters to a chart and look for clusters of optimum parameters. Ie: To make sure I didn't just get lucky with a stop of XYZ.

Still working to replace the % trailing for a supertrend / ATR based trailing stop. Its hard to implement but I have a feeling going to be worth it!

Reply With Quote
 
(login for full post details)
  #340 (permalink)
 RM99 
Austin, TX
 
Experience: Advanced
Platform: TradeStation
Trading: Futures
 
Posts: 882 since Mar 2011
Thanks: 124 given, 701 received


andyb1979 View Post
Thanks!

Ok good so its not impossible then (to create a working trading strategy). Yes from speaking to several traders I have been told to take a percentage of profit early and move a stop to break even. Only when I implemented this in software and simulated it did I see how powerful it actually is.

re: Optimisation and curve fitting, I get it. I feel more confident that I did not touch the entry rules, just exit, that this has a chance of working live. Also I plan when I optimise to output the optimisation parameters to a chart and look for clusters of optimum parameters. Ie: To make sure I didn't just get lucky with a stop of XYZ.

Still working to replace the % trailing for a supertrend / ATR based trailing stop. Its hard to implement but I have a feeling going to be worth it!

Some programs will actually do a robustness analysis for you. But essentially, that's what you're manually doing by picking settings that are resistant to large swings in performance.

At some point, when I get sophisticated enough, I'm going to develop multi dimensional mapping that will show "heat maps" of where particular nodes of high performance ocurr. These maps allow you to visually select plateau areas and avoid very narrow spikes (where the performance falls off sharply with small adjustment). Basically, it's a robustness check in visual form.

"A dumb man never learns. A smart man learns from his own failure and success. But a wise man learns from the failure and success of others."
Reply With Quote
 
(login for full post details)
  #341 (permalink)
 RM99 
Austin, TX
 
Experience: Advanced
Platform: TradeStation
Trading: Futures
 
Posts: 882 since Mar 2011
Thanks: 124 given, 701 received

I must caution you though, that if your backtesting is inherently flawed, all of this will be for nothing.

I see you went back to daily bars. Just bear in mind that if you're not backtesting tick/tick (or with market replay as the Ninjas call it) then again, if your transaction can ocurr all within the same bar, the engine has no way of knowing what ocurred inside that bar and what direction the bar moved first, in the middle, etc.

All of this optimization and what not has to assume there are no fatal flaws in your backtesting method and ability to translate forward.

At a bare minimum you should do a sanity check and forward test for some sample size, just to verify everything looks as good going forward as it did backward.

"A dumb man never learns. A smart man learns from his own failure and success. But a wise man learns from the failure and success of others."
Reply With Quote
 
(login for full post details)
  #342 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Im using OpenQuant (trial) as it supports portfolio rebalancing and other features that I want to use. It doesn't unfortunately do sophisticated analysis of optimization parameters (although I am writing code to spit optimization out to file so I can view the results in Excel). It does do robustness as you can create a Monte Carlo instrument (simulated stock price) using the existing instrument drift (trend), volatility etc... Not sure how this works but I will certainly be investigating it.

Reply With Quote
 
(login for full post details)
  #343 (permalink)
 TheTrend 
Paris, France
 
Experience: Advanced
Platform: Amibroker
Trading: Futures, Stocks
 
Posts: 96 since Jun 2011
Thanks: 118 given, 103 received

I must second RM99 and let you know that your result might be flawed because your take profit of 28 pips seems really small compared to an average daily range of 110pips for EUR/USD.

If you consider your signal to be more reliable based on the daily time frame, take your signals based on that but manage and test your stratey on a lower time frame. You can do that quite easily in MultiCharts, don't know about Open Quant though.

Good luck with your strategy !

Reply With Quote
 
(login for full post details)
  #344 (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: 50,002 since Jun 2009
Thanks: 32,468 given, 98,268 received


andyb1979 View Post
Ok, I took a step back and applied a trend following strategy I've developed on the daily EURUSD only. Forget 15 mins for now, its too noisy. I need to go back to basics and back to the daily chart.

I suggest you create your own journal thread on the forum to track everything. And more importantly, at the end of every week, go back and read all the journal posts for the prior two weeks. Hopefully it will help you avoid what I call the black hole phenomenon, where traders repeat the same mistakes over and over again without realizing it.

Here is an excellent example of a systems "journal". The author is publishing all of his work and code, but the important lessons are what happens over time and how he and the system are evolving:



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)
  #345 (permalink)
 Jigsaw Trading  Jigsaw Trading is an official Site Sponsor
Site Sponsor

Web: Jigsaw Trading
AMA: Ask Me Anything
Webinars: Jigsaw Trading Webinars
Elite offer: Click here
 
 
Jigsaw Trading's Avatar
 
Posts: 2,982 since Nov 2010
Thanks: 825 given, 10,323 received


andyb1979 View Post
Here are (honest) answers to your questions

Include commissions - No (Openquant doesnt support this in test mode, as far as I can tell)
Slippage - No
Orders - All at market on bar close
Expectancy - with these settings about +$35 per trade.

i.e. average of all losses + all wins / total trades is $35

So I know what you're thinking. $35 - some slippage - some commissions (about $3 per side on interactive brokers FX) = potential to go negative in live trading.

Well, its a start

I'd like to see the expectancy higher. The win:loss ratio Im pleased with at its 68%. The average win however is not far off the average loss.

I'm trying to add some better profit taking like scalp 50% at a certain level then trail the rest to see how that improves expectancy

Isn't this an unrealistic system if it expects to be filled at the closing price of a bar?

Visit my futures io Trade Journal Reply With Quote
 
(login for full post details)
  #346 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

In the lab for a few days, this is what I've come up with.

A trend/scalp hybrid strategy - basically trades the trend but also takes profit at 2 points (using ATR - volatillity estimate - for profit targets).

The entry is fixed but exit I've optimised based on ATR window and multiplier to take profit at 2 points. Thats it. I've tested on a basket of stocks, ETFs and currencies (need more data to test more instruments), including commissions and slippage. max drawdown is around 15% in the credit crunch. Its a long only strategy.

Next up, opening a futures account & paper trading

Attached Thumbnails
Click image for larger version

Name:	Trend strat result.png
Views:	133
Size:	223.1 KB
ID:	48129  
Reply With Quote
 
(login for full post details)
  #347 (permalink)
 GT22 
Currently Inactive
 
 
Posts: 25 since Aug 2011

@andyb1979

Suggestion:

Eventually you want will want to move into strategies that are adaptable both horizontally and vertically on the charts. Do that and the backtesting becomes less important. ATR is crude and is a start.

Reply With Quote
 
(login for full post details)
  #348 (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: 50,002 since Jun 2009
Thanks: 32,468 given, 98,268 received

Moderator Notice
Moderator Notice



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)
  #349 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Sorry

... So thanks to rigourous simulation & backtest and a nice pretty equity curve, I now feel confident to pull the trigger. Whereas If I'm left to make the decisions I'll bottle it and make a right pigs ear of it

Reply With Quote
 
(login for full post details)
  #350 (permalink)
 forrestang 
Urban Samurai
Chicago IL
 
Experience: None
Platform: Ninja, MT4, Matlab
Broker: CQG, AMP, MB, DTN
Trading: E/U, G/U
 
forrestang's Avatar
 
Posts: 1,227 since Jun 2010
Thanks: 302 given, 929 received


DionysusToast View Post
Isn't this an unrealistic system if it expects to be filled at the closing price of a bar?

Some stuff I was messing with yesterday. Pertaining to your comment above, I noticed a similar phenomenon. I was using an exotic bar type, and doing some crude backtesting with NT7. The fills were unlikely, VERIFIED UNREALISTIC with a replay.

I guess the cool thing was how consistent the system was. No matter the size of the bar, chart resolution or the period in which I tested (tested 3 months at a time for the whole year of 2011). All equity curves had almost exact same shape, winning % and everything remained consistent. Increasing the bar size would linearly increase the expectancy per trade.

I believe the pics below where of 6E.... oil was really insane. I deleted most of the stats, here where just two examples using a really small bar size with 1 contract.

Attached Thumbnails
Click image for larger version

Name:	Prime2011-09-01_080132.jpg
Views:	118
Size:	191.3 KB
ID:	48161   Click image for larger version

Name:	Prime2011-09-01_082213.jpg
Views:	111
Size:	193.7 KB
ID:	48162  
Reply With Quote
 
(login for full post details)
  #351 (permalink)
 Xeno 
UK
 
Experience: Intermediate
Platform: Ninja
Broker: Mirus/Zen
Trading: Futures - bonds, currencies, index
 
Posts: 288 since Oct 2010
Thanks: 70 given, 270 received


Xeno View Post

In my backtesting career, I have had about four occassions where a very good equity curve and stats jumped out at me early in the testing process. On 100% of those occassions there was an error in the strategy or backtesting process. By the fourth time it happened, there was no excitement. I knew immediately what the likely explanation was. I'm afraid that's the sort of sceptical mindset you have to have.

And by concidence, my latest scalping strategy just generated a perfect example, which I've attached. It's not a bad one, since there are few real red flags on it - only how succesful it is, and the large inbalance between short/long.

Attached Thumbnails
Click image for larger version

Name:	NinjaTrader Performance Report.jpg
Views:	192
Size:	118.7 KB
ID:	49298  
Reply With Quote
 
(login for full post details)
  #352 (permalink)
andyb1979
London UK
 
 
Posts: 37 since Aug 2011
Thanks: 7 given, 4 received

Rofl yes 300k short and far less long. Just out of interest what was the input time series? If a rabid bear market then it explains the results. However I'd test it on bull market and paper trade to confirm!

Reply With Quote
 
(login for full post details)
  #353 (permalink)
 Xeno 
UK
 
Experience: Intermediate
Platform: Ninja
Broker: Mirus/Zen
Trading: Futures - bonds, currencies, index
 
Posts: 288 since Oct 2010
Thanks: 70 given, 270 received

Range bars 5. It was a runaway strategy that kept opening positions in a way that wouldn't work live. As for bull and bear - it was all one day.

Reply With Quote
 
(login for full post details)
  #354 (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: 50,002 since Jun 2009
Thanks: 32,468 given, 98,268 received



A lot of specific information on how Peter Campbell's M3 fund auto trading systems work, how he designs them, ideas behind them etc.

Peter will be here on Thursday @ 4:30PM for a webinar:



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)
  #355 (permalink)
 bluemele 
Honolulu, Hawaii
 
Experience: Intermediate
Platform: NinjaTrader
Broker: ATC/TT, AMP/Zen-Fire, AMP/CQG
Trading: TF
 
bluemele's Avatar
 
Posts: 2,547 since Jun 2010
Thanks: 3,803 given, 2,836 received

I ran a similar auto strategy from the looks of it. Seems like he is using martingale or averaging in or whatever you want to call it. My system worked very well as well, but I didn't like 20% drawdowns as I was just trading too much of my account. If you trade 10% of your account like he talks about, then no reason you can't pull down 2%3% easy a month which most people would kill for if you are managing a large amount.

Visit my futures io Trade Journal Reply With Quote
 
(login for full post details)
  #356 (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: 50,002 since Jun 2009
Thanks: 32,468 given, 98,268 received


bluemele View Post
I ran a similar auto strategy from the looks of it. Seems like he is using martingale or averaging in or whatever you want to call it. My system worked very well as well, but I didn't like 20% drawdowns as I was just trading too much of my account. If you trade 10% of your account like he talks about, then no reason you can't pull down 2%3% easy a month which most people would kill for if you are managing a large amount.

Would you mind re-posting that in the Webinar thread? I have more to say, but want to keep it in that thread and not this one - since this one is supposed to be about psychology and not methodology.

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)
  #357 (permalink)
 GaryD 
Orlando, Florida
 
Experience: None
Platform: shoes
Trading: happy
 
GaryD's Avatar
 
Posts: 6,462 since May 2011


liquidcci View Post
I tried just about every mind trick I could to get my trading right. But found I was fatally flawed. Trading seemed to bring out my worst characteristics. I would read my trading plan and my rules out loud every day before market opened. I would type up contracts with myself that I had to sign but none of it worked.

One time I did over 100 emini trades in a day 1 contract at a time only ending when I blew up my account. I was not scalping but just out of control. I was in a cold sweat and I could not stop pushing that entry button. The trading ladder jumping up and down could just not be resisted. Like a carnival game that could not be won I went until all my money was gone. I ended that day in a fetal position in the corner with my thumb in my mouth. It was bad. Trading reduced me to a slot machine gambler with no discipline.

But then I decided to build my signals into an auto trade strategy. It was the best time and money I ever spent. It 100% solved my discipline issues. I have not entered a rogue trade outside of my plan for 2 years and am profitable. This may not work for everyone but was only way I could overcome the darkness that lurks in every trader.

Hey liquidcci,

I realize this was your first post in this thread, sorry. I just happened to have this pop up when I searched for how to define specifc times of day in NT.

When I started trading I had $1,000.00 a day commissions some times. I have blown up accounts. I have made rules and broken them. I found your post very humble and real. Trading showed me my worst qualities, over and over and over, and I'm sure it does that to a lot of us. I wrote several dozen systems that I tried live, but as strong as my temptation to break my own rules was, my inability to walk away and let the computer decide was far stronger. Maybe I have control issues, or fear, or some other defect. So, if I would not listen to myself, and would not trust a robot, it seemed like I should stop trading.

But, I could not end the obsession. I came into trading having made a lot of money for the majority of my life, but then lost a tremendous amount. I saw trading as something to toy with, but fairly quickly it presented itself as a way to get back where I had been for so long. Not that any of us will ever harness anything near this, but the potential is there to double your money, daily. The key drug in trading is "potential". And as habit forming as any other.

I decided to change myself, and still am on that path. I started eating better, re-joined the gym, and sim traded non-stop. Played with colors, indicators, backgrounds, screen layout, markets, read every book I could find on trading and set my charts up to that new recommended method, then scrapped it and started over...Same obsesion, re-directed. Sim, sim, sim, sim. Traded $2,500.00 sim accounts, beside $10,000,000.00 sim acounts, and reset them often. Named them sometimes even. Had names for my stop and profit configurations, and for my setups. Over, and over and over. Until I was numb to it. I really did not care which way a trade went. What did it matter? It was simulated. And, I had already proven I could make or loose riduculous amounts in my fictitious world. And finally, I was calm.

The "drug" is always there, and once an addict we are possibly always suspect. But, ______oholics can recover. Trading takes a lot of work, and as was so pure in your opening line, the overwhelming majority of it has nothing to do with the markets. Thanks for your post.

Reply With Quote
 
(login for full post details)
  #358 (permalink)
 liquidcci 
Austin, TX
 
Experience: Master
Platform: ninjatrader, r-trader
Trading: NQ, CL
 
liquidcci's Avatar
 
Posts: 865 since Jun 2011
Thanks: 610 given, 1,070 received


GaryD View Post
Hey liquidcci,

I realize this was your first post in this thread, sorry. I just happened to have this pop up when I searched for how to define specifc times of day in NT.

When I started trading I had $1,000.00 a day commissions some times. I have blown up accounts. I have made rules and broken them. I found your post very humble and real. Trading showed me my worst qualities, over and over and over, and I'm sure it does that to a lot of us. I wrote several dozen systems that I tried live, but as strong as my temptation to break my own rules was, my inability to walk away and let the computer decide was far stronger. Maybe I have control issues, or fear, or some other defect. So, if I would not listen to myself, and would not trust a robot, it seemed like I should stop trading.

But, I could not end the obsession. I came into trading having made a lot of money for the majority of my life, but then lost a tremendous amount. I saw trading as something to toy with, but fairly quickly it presented itself as a way to get back where I had been for so long. Not that any of us will ever harness anything near this, but the potential is there to double your money, daily. The key drug in trading is "potential". And as habit forming as any other.

I decided to change myself, and still am on that path. I started eating better, re-joined the gym, and sim traded non-stop. Played with colors, indicators, backgrounds, screen layout, markets, read every book I could find on trading and set my charts up to that new recommended method, then scrapped it and started over...Same obsesion, re-directed. Sim, sim, sim, sim. Traded $2,500.00 sim accounts, beside $10,000,000.00 sim acounts, and reset them often. Named them sometimes even. Had names for my stop and profit configurations, and for my setups. Over, and over and over. Until I was numb to it. I really did not care which way a trade went. What did it matter? It was simulated. And, I had already proven I could make or loose riduculous amounts in my fictitious world. And finally, I was calm.

The "drug" is always there, and once an addict we are possibly always suspect. But, ______oholics can recover. Trading takes a lot of work, and as was so pure in your opening line, the overwhelming majority of it has nothing to do with the markets. Thanks for your post.

You said it so well. There is a very dark addictive side to trading. It can be beat but is a hard road because it involves going against our own nature. Thanks for your comments.

"The day I became a winning trader was the day it became boring. Daily losses no longer bother me and daily wins no longer excited me. Took years of pain and busting a few accounts before finally got my mind right. I survived the darkness within and now just chillax and let my black box do the work."
Started this thread Reply With Quote
 
(login for full post details)
  #359 (permalink)
 xplorer 
Site Moderator
London UK
 
Experience: Beginner
Platform: CQG
Broker: S5
Trading: Futures
 
xplorer's Avatar
 
Posts: 5,349 since Sep 2015
Thanks: 13,484 given, 12,939 received


rtrade View Post
Hmm, I if I understand the video, the girl is on RM99 and liquidCCI's side and the boy is on DT's side.
But I'm not sure, you guys can decide.

[yt]http://www.youtube.com/watch?v=_futures.io (formerly BMT)6BfxR7w[/yt]

I don't have anything of value to contribute to this thread - it's pretty clear where everybody's at.


But I absolutely had to repost the correct link to the video of the 2 sides bickering (which had BMT in the URL and was hence caught in the global search and replace), which is hilarious.


Reply With Quote


futures io Trading Community Psychology and Money Management > Auto trading is only thing that conquered my darkside


Last Updated on January 17, 2016


Upcoming Webinars and Events
 

NinjaTrader Indicator Challenge!

Ongoing
 

Journal Challenge w/$1,800 in prizes!

April
     



Copyright © 2021 by futures io, s.a., Av Ricardo J. Alfaro, Century Tower, Panama, +507 833-9432, 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