NexusFi: Find Your Edge


Home Menu

 





Want to create an indicator and have no idea where to start?


Discussion in NinjaTrader

Updated
      Top Posters
    1. looks_one forrestang with 58 posts (9 thanks)
    2. looks_two ThatManFromTexas with 16 posts (5 thanks)
    3. looks_3 LostTrader with 6 posts (2 thanks)
    4. looks_4 cory with 5 posts (3 thanks)
      Best Posters
    1. looks_one cory with 0.6 thanks per post
    2. looks_two ThatManFromTexas with 0.3 thanks per post
    3. looks_3 LostTrader with 0.3 thanks per post
    4. looks_4 forrestang with 0.2 thanks per post
    1. trending_up 25,707 views
    2. thumb_up 20 thanks given
    3. group 5 followers
    1. forum 93 posts
    2. attach_file 30 attachments




 
Search this Thread

Want to create an indicator and have no idea where to start?

  #71 (permalink)
 
forrestang's Avatar
 forrestang 
Chicago IL
 
Experience: None
Platform: Ninja, MT4, Matlab
Broker: CQG, AMP, MB, DTN
Trading: E/U, G/U
Posts: 1,329 since Jun 2010
Thanks Given: 354
Thanks Received: 1,047


lolu View Post
When you post the final release, could do two versions; one as an indicator and the other as your strat version.

Lolu

Yep. That's the plan.

Can backtesting only be done via strat?

Started this thread Reply With Quote

Can you help answer these questions
from other members on NexusFi?
The space time continuum and the dynamics of a financial …
Emini and Emicro Index
Exit Strategy
NinjaTrader
Deepmoney LLM
Elite Quantitative GenAI/LLM
NexusFi Journal Challenge - April 2024
Feedback and Announcements
New Micros: Ultra 10-Year & Ultra T-Bond -- Live Now
Treasury Notes and Bonds
 
Best Threads (Most Thanked)
in the last 7 days on NexusFi
Get funded firms 2023/2024 - Any recommendations or word …
59 thanks
Funded Trader platforms
36 thanks
NexusFi site changelog and issues/problem reporting
25 thanks
GFIs1 1 DAX trade per day journal
19 thanks
The Program
18 thanks
  #72 (permalink)
 
lolu's Avatar
 lolu 
Lagos, Nigeria
Market Wizard
 
Experience: Intermediate
Platform: NinjaTrader, SierraChart
Trading: Euro Currency & Oil
Frequency: Daily
Duration: Hours
Posts: 2,552 since Jun 2009
Thanks Given: 1,049
Thanks Received: 1,678


forrestang View Post
Yep. That's the plan.

Can backtesting only be done via strat?

Any method or setup can be backtested. A strat is a method, usually automated to perform a defined process.

Lolu

Visit my NexusFi Trade Journal Reply With Quote
  #73 (permalink)
 
ThatManFromTexas's Avatar
 ThatManFromTexas 
Houston,Tx
 
Experience: Advanced
Platform: NinjaTrader
Broker: Mirus Futures/Zen-Fire
Trading: TF
Posts: 2,265 since Feb 2010
Thanks Given: 1,206
Thanks Received: 4,348



forrestang View Post
Yep. That's the plan.

Can backtesting only be done via strat?


I have never found a connection between back testing and actual results.

When you "gitterdun" ... run it in sim for a while see how it does.

I'm just a simple man trading a simple plan.

My daddy always said, "Every day above ground is a good day!"
Reply With Quote
  #74 (permalink)
 
forrestang's Avatar
 forrestang 
Chicago IL
 
Experience: None
Platform: Ninja, MT4, Matlab
Broker: CQG, AMP, MB, DTN
Trading: E/U, G/U
Posts: 1,329 since Jun 2010
Thanks Given: 354
Thanks Received: 1,047


ThatManFromTexas View Post
I have never found a connection between back testing and actual results.

When you "gitterdun" ... run it in sim for a while see how it does.

Oh ya, I know. It's usually not realistic.

I'm looking to more so collect stats than anything though. Just a general idea to test some theories about depth of pb and continuation etc.

This thread is probably weird as it's dis-jointed, and really just a working journal of the creation of this thing, a place to think out loud I suppose. I didn't do a good job explaining what I was looking to capture with this info.

Started this thread Reply With Quote
  #75 (permalink)
 
ThatManFromTexas's Avatar
 ThatManFromTexas 
Houston,Tx
 
Experience: Advanced
Platform: NinjaTrader
Broker: Mirus Futures/Zen-Fire
Trading: TF
Posts: 2,265 since Feb 2010
Thanks Given: 1,206
Thanks Received: 4,348


forrestang View Post
Oh ya, I know. It's usually not realistic.

I'm looking to more so collect stats than anything though. Just a general idea to test some theories about depth of pb and continuation etc.

This thread is probably weird as it's dis-jointed, and really just a working journal of the creation of this thing, a place to think out loud I suppose. I didn't do a good job explaining what I was looking to capture with this info.

But if the statistics Back Testing gives you is flawed... what good is it?

I'm just a simple man trading a simple plan.

My daddy always said, "Every day above ground is a good day!"
Reply With Quote
  #76 (permalink)
 
forrestang's Avatar
 forrestang 
Chicago IL
 
Experience: None
Platform: Ninja, MT4, Matlab
Broker: CQG, AMP, MB, DTN
Trading: E/U, G/U
Posts: 1,329 since Jun 2010
Thanks Given: 354
Thanks Received: 1,047


ThatManFromTexas View Post
But if the statistics Back Testing gives you is flawed... what good is it?

Good question. I have some general ideas, based on me trading this way for a while, without the visual aid I've got now. I've got quite a few statistics I've been keeping for a while, but some things I just don't have much data for. So I'm hoping that if the expectancies on back vs. forward testing holds, it could be useful?

With all the stats I have collected via walking forward for a while now, there appears to be a 'general expectancy' associated with certain trade areas. I'm curious to see if the stats I have collected, compare in anyway to the testing I will do. Even if they don't line up, I'd like to see if the proportionality holds.

For example, using arbitrary numbers and pretend trades.......
Say I had two setups, ABC and XYZ. Say my forward testing showed me ABC yielded an expectancy of about $5.00/trade. And then say XYZ showed an expectancy of $25.00/trade.

Then say I wrote a script to backtest these two strats, and say ABC yielded $10.00/trade, which is double the forward testing. And maybe XYZ yielded $50.00/trade. Hypothetical of course.

So the results are completely unrealistic, but if the expectancy holds proportionally...... I wouldn't call the back testing realistic, but it could be useful.

I have some trade areas that I have been forward testing, that just don't perform that well. They ALL appear to be profitable, but some trade areas just seem to blow others away and seem to have a much better expectancy, which if is the case, I can widdle down my trade selection, and just leverage the better RR trades across more than one instrument.

But of course, the trades that are doing the best, i have less of them. So I'm wondering if some trades are just better, or if some trades just seem better because they have traded less, and just haven't tapered off to their normal expectancy yet, in the way the less profitable trades already have since there are more of them.

I could be off on my logic of course.

Started this thread Reply With Quote
  #77 (permalink)
 
cory's Avatar
 cory 
virginia
 
Experience: Intermediate
Platform: ninja
Trading: NQ
Posts: 6,098 since Jun 2009
Thanks Given: 877
Thanks Received: 8,090

I watch it in real time the triangle signals move when there is new high, new low, the confirm signal with words arrives much latter waiting for confirmation I guess. You may want to further filter black triangles with my pink/green signal

Reply With Quote
Thanked by:
  #78 (permalink)
 
forrestang's Avatar
 forrestang 
Chicago IL
 
Experience: None
Platform: Ninja, MT4, Matlab
Broker: CQG, AMP, MB, DTN
Trading: E/U, G/U
Posts: 1,329 since Jun 2010
Thanks Given: 354
Thanks Received: 1,047


cory View Post
I watch it in real time the triangle signals move when there is new high, new low, the confirm signal with words arrives much latter waiting for confirmation I guess. You may want to further filter black triangles with my pink/green signal

For some reason I didn't get an email of this post. Could be just what I need. I will check it out!!

Started this thread Reply With Quote
  #79 (permalink)
 
forrestang's Avatar
 forrestang 
Chicago IL
 
Experience: None
Platform: Ninja, MT4, Matlab
Broker: CQG, AMP, MB, DTN
Trading: E/U, G/U
Posts: 1,329 since Jun 2010
Thanks Given: 354
Thanks Received: 1,047

On another note, I got the strat working a while back, it wasn't much to it actually. Everything is still in pieces, but it works.

I've been having it running and walking it forward this week, and it is doing pretty nicely. I like the way the betterRenko bars are spiking my oscillations. When you look at the chart, they are much more spikey, they seem to highlight the oscillations nicer than normal RangeBar charts do. But all that is still up for debate.

This week, while being decent so far, if you look at my chart, the short term strength on the bottom is mostly UP, while my Long term(color bars) have been in conflict. So the trades aren't killer. There is a slightly better expectancy when they match, price usually moves farther/faster.

So far, this week, my win% is hovering right around 40%, which is what I've seen in my manual trading so far. My Average Win/Loss is still holding at about a bit better than 2:1.

After lunch today, here is the chart of the week. I've removed the actual indicator itself that plots the arrows, as the strat and study will plot slightly differently. It's pretty wide.




// Silliness Below-----------------------------------------------------------------------------------------------------
Just for fun..... was messing around collecting stats on a few different continuous contracts, and came up on a horrible combination. It had a pretty high win% relative to the things I've been doing, but a low expectancy of $9.22/trade. I was just curious and bored, so I made another strat that was the binary inversion of the original strat, so it was told to buy when the original said sell and vice versa.

Here is the original horrible strat and curve:


Here is the inverted version:

Started this thread Reply With Quote
Thanked by:
  #80 (permalink)
 
forrestang's Avatar
 forrestang 
Chicago IL
 
Experience: None
Platform: Ninja, MT4, Matlab
Broker: CQG, AMP, MB, DTN
Trading: E/U, G/U
Posts: 1,329 since Jun 2010
Thanks Given: 354
Thanks Received: 1,047



cory View Post
I watch it in real time the triangle signals move when there is new high, new low, the confirm signal with words arrives much latter waiting for confirmation I guess. You may want to further filter black triangles with my pink/green signal

Just for reference, those arrows that show up on the original study I posted, they aren't anything to trade off of. They're just like "show-me" dots in that they just plot highs/lows. Then there is additional logic to tell if it's a HH/LL/HL or LH. And it does that by taking the highest, lowest point that occurred when that oscillator changes colors.

Once it has those points, it defines trend via...... L-H-HL-HH means that we are in an uptrend. H-L-LH-LL means we are in downtrend. Breaking a HL or LH transitions to trend change etc...

Started this thread Reply With Quote




Last Updated on January 20, 2012


© 2024 NexusFi™, s.a., All Rights Reserved.
Av Ricardo J. Alfaro, Century Tower, Panama City, Panama, Ph: +507 833-9432 (Panama and Intl), +1 888-312-3001 (USA and Canada)
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.
About Us - Contact Us - Site Rules, Acceptable Use, and Terms and Conditions - Privacy Policy - Downloads - Top
no new posts