Let's Compare Strategies

[NFT-Trader]

Replay Offer

Sledge,



in fact, it is possible. I do this all the time. For example download the various contracts you want to replay, lets take 03-12 and 06-12. Go to your NT database directory(C:\Users\*****\Documents\NinjaTrader 7\db\data ) MAKE A BACKUP of the entire directory, and rename all the "ES 06-12.ntm" to "ES 03-12.ntm" with a file renamer program (I use "filerenamer", it takes just a few clicks). Now you have a continuous contract over 6 months.


NFT Trader,



I have a continious ES contract from june 2011 to june 2012 in my database, I'd be more than happy to do a replay for you

Thanks for the offer Marcow but I am going to decline it.

[NFT-Trader]

Download Script

I have a script to auto download market replay data from within NT.

I am going to start working on one now that you say it can be done.

[NFT-Trader]

Walking Forward

Walk forward results are the only results that count, otherwise a simple optimization is just predicting historical data perfectly (especially when you >3 parameters, you mentioned you have 50!). Either way, looks like it's off to a good start, but 1 month of live results may not be enough, unless it's profiting every day.

I appreciate the advice on walking forward and I will try to put more emphasis into using it as I better understand its value. However, if I am optimizing, lets say from 11/1/11 to 11/30/12 to be used for Dec 12 then am I not in good shape (albeit no one can predict the future) if for the last 12 months I am end up with the numbers I presented below. I need to understand how walking forward is more valuable than consistent profits over a long period. Originally I tried optimizing for 3 weeks but the results were skewed in that I can easily curve fit my parameters to any given 3-4 weeks.

I am going to play with walking forward and hopefully post some results.Again thanks for the advice.

Other than Sledge I still do not see anyone post their numbers so that we can compare and have this educational conversation.

[:::grimReaper:::]

Walking forward results have been as good as optimize so far. I am not sure I have a grasp on the value of walking forward yet..

Walk forward results are the only results that count, otherwise a simple optimization is just predicting historical data perfectly (especially when you >3 parameters, you mentioned you have 50!). Either way, looks like it's off to a good start, but 1 month of live results may not be enough, unless it's profiting every day.

[sledge]

Walking forward results have been as good as optimize so far. I am not sure I have a grasp on the value of walking forward yet.

I have live traded a version of this strategy since October and the account is up 45% and results have been within the variation of the strategy.

Not sure how to run replay mode for the entire year since I think I need to download each day individually. Is this correct? My system is very fast so I should be good to go once the data is on hand.

I have about 700 lines of code and ~ 50 variables.

I have a script to auto download market replay data from within NT.

[marcow]

Sledge,

I'm not sure how to run a full market replay across contracts, I don't think that is possible.

in fact, it is possible. I do this all the time. For example download the various contracts you want to replay, lets take 03-12 and 06-12. Go to your NT database directory(C:\Users\*****\Documents\NinjaTrader 7\db\data ) MAKE A BACKUP of the entire directory, and rename all the "ES 06-12.ntm" to "ES 03-12.ntm" with a file renamer program (I use "filerenamer", it takes just a few clicks). Now you have a continuous contract over 6 months.


NFT Trader,

Not sure how to run replay mode for the entire year since I think I need to download each day individually. Is this correct? My system is very fast so I should be good to go once the data is on hand.

I have a continious ES contract from june 2011 to june 2012 in my database, I'd be more than happy to do a replay for you

[NFT-Trader]

NFT, did you try walk-forward? How many parameters does your strategy have?

Walking forward results have been as good as optimize so far. I am not sure I have a grasp on the value of walking forward yet.

I have live traded a version of this strategy since October and the account is up 45% and results have been within the variation of the strategy.

Not sure how to run replay mode for the entire year since I think I need to download each day individually. Is this correct? My system is very fast so I should be good to go once the data is on hand.

I have about 700 lines of code and ~ 50 variables.

[:::grimReaper:::]

NFT, did you try walk-forward? How many parameters does your strategy have?

[koganam]

That 09/2012 ES contract

I got it to

total +7587, long 7762, short -175

so i lost some from the long, but gained more from the short with something i saw.

At my slow processor speed, it will take me a few more nights/days to test in Market Replay ES 03-12, and ES 06-12. And I will need to rerun the code through 12-12 (although I don't think any of the conditions that were in 09-12 happened in 12-12 yet).

I should be able to provide market replay results for 2012 (summed together).

I'm not sure how to run a full market replay across contracts, I don't think that is possible.

Well, I have a much faster system. If I can see the actual Strategy code, I can give it a run. The results so far look impressive.

[sledge]

Any chance that we can see the strategy code. The results look pretty impressive.

That 09/2012 ES contract

I got it to

total +7587, long 7762, short -175

so i lost some from the long, but gained more from the short with something i saw.

At my slow processor speed, it will take me a few more nights/days to test in Market Replay ES 03-12, and ES 06-12. And I will need to rerun the code through 12-12 (although I don't think any of the conditions that were in 09-12 happened in 12-12 yet).

I should be able to provide market replay results for 2012 (summed together).

I'm not sure how to run a full market replay across contracts, I don't think that is possible.

[sledge]

Sledge has encouraged me to post my screen shots. This is the Strategy I will be live trading going into December. Hope it holds for the month.

Is this with Market Replay or just Strategy Analyzer?

You should verify in Market Replay....

If you can't run market replay, at least run in SIM mode live on it before you go for "real" money.

[sledge]

Hey Sledge, Thanks for the reply and the numbers. Have you tried seeing whether it is consistent over a longer period. I find with my strategy I can "curve fit" it very nicely over a few months but extending that over a year never seems to work. Therefore I have decided, without any support for my logic yet, that I will live trade my system after optimizing it over the year.

By the way what is a floater?

Hi

Oops, floater was a word I made up at the moment.

To be honest, that final 1 contract is at a fixed target. I haven't applied any ATM or moving stop less strategy to see if I can get more out of it if a "runner" happens.

In my case case, a 3:1 ratio looks doable. I could scale into 6:2 if that is successful. I have not tested 4:0 ratio yet. If 4:0 works out, then 8:0 would work better

It took about 6-7 hours on a quad core non hyper thread Intel Win Xp 32 to do 2 charts for market replay at 500x for full 09-2012 ES contract. There was 0 blip across cores from NT and memory was at 800 megs the whole time.

My dual core AMD athlon 4400+ X2 (circa 2006), it running at 50% on both cores, with about 10 charts up. I think I would expect 12-16 (24?) hours to complete at this rate.

My # of trades appear to be confusing in NT, as # of contracts would be more correct or more valid in the results.

I have "max entries per direction = 2".. and I enter "Trade1" at 3 contracts, and "Trade2" at 1 contract.

I verified that the 1st day, show "2 trades".. even though it was skewed 3:1 .

[koganam]

ES 09-12 market replay attached as images.

I only made printouts of ES 12-12 market replay test (up until black friday):

total net profit: 7,550
gross profit 11,162
gross loss -3,612

long total net: 4712
long profit: 5962
long loss : 1250

short total net 2837
short profit: 5200
short loss: 2362

sharpe total: 1.45

date 9/17/2012 to 11/29/2012

totale # of trades: 59, long 33, short 26
percent profitable: 73%, 79%, 65%

ratio win/loss : total 1.21, long 1.32, short 1.21

max drawdown total -1.01%, long -0.65%, short -0.91%


both were 4 contracts, 2 entries, 3 contracts at a fixed target, and 1 floater.

I'm not sure hot total # of trades is computed yet, I haven't verified, so I can't verify commission of $7 round trip.

no optimization attempted yet.

Do I have guts to play this live yet? NOPE

Any chance that we can see the strategy code. The results look pretty impressive.

[koganam]

Sledge has encouraged me to post my screen shots. This is the Strategy I will be live trading going into December. Hope it holds for the month.

Any chance that we can see the strategy code. The results look pretty impressive.

[NFT-Trader]

Screen Shots of Strategy

Sledge has encouraged me to post my screen shots. This is the Strategy I will be live trading going into December. Hope it holds for the month.