Proper back and forward testing

Discussion in 'Automated Trading' started by Allistah, May 4, 2013.

  1. Allistah


    Hi there..

    I hope this is the right forum for this question.

    Lets say that my strategy runs on a 5m chart and it only trades during regular trading session hours.

    How do you properly back and forward test this? Optimize on 4 months and then forward test the 5th month? The move up to the next 4 months that include the prior forward gest and optimize that. Then forward again the next month forward.

    How do I know what the right thing to start off with is? Is there any good books that can teach you about proper back and forward testing?


  2. Whatever you do, realize that- brutally speaking - no retail software will properly help you with that ;)
  3. The Evaluation and Optimization of Trading Strategies book is a good start. But as NetTecture said, you definitely will need to write your own code to do the testing thoroughly.
  4. ofthomas


    the only retail tool i know will do this, close enough... is neuroshell..

  5. Yes, that is the only one. Intresting enough it is not a "major" one - as in: not one you see people talking all the time. Likely due to them not stuffing it down any beginner's throat with a free version.

    I am not sure sure they do backtest with full order book, though ;)

    It was the only tool standing when we did our evaluation - at the end we decided to go with an inhouse solution (thanks a partner now had developped a VERY good start) due to (a) C#, (b) Visual Studio and (C) the ability to just update it ;) we do regular updates at the moment, rolling out ike 2 new versions a week - even more of the UI when I get a request for something one of my dudes needs ;)
  6. dom993


    You guys really make me laugh.

    Obviously the OP knows nothing about automated trading, and your best advice to him is for him to develop his own trading platform?!

    OP, my advice to you is to download either NinjaTrader free or Multichart .NET starter, and use either of these plateform to discover the world of automated trading. Of course, you'll need quality historical data, and for that I would send you visit TickData. For about $1300 you'll get 30 year.symbol, for example 10 years of ES + 10 years of 6E + 10 years of CL. This is an expense you cannot skip, else just find something else to do than attempting automated trading.

    IF at some point you have enough experience, knowledge & reasons to develop your own platform, you'll cross that bridge. FWIW, I trade 100% automated with Ninja, my primary system is always in the market (24/7/365, week-ends included), and aside from starting the system every Sunday afternoon I do NOTHING - it is 100% automated, including recovery for all kind of events, such as loss of connectivity (and thanks to IB, this happens every night). That system had a good week, +6450 for 16 trades (1 contract), and I couldn't care less of the order book in backtesting.

    I am not saying Ninja is perfect out of the box, far from it, but it did cost me a lot less to develop the workarounds that my trading systems require, instead of building a platform from the ground up. In general I do not use optimization, except for comparing trade management options (targets & stop).

    I know I already shared that backtest on ET, anyway this is my primary system, trading 1 contract on every signal, without any limit to the position size (at times it goes beyond 10 contract, but the average position is 5 contracts). That performance level is WITHOUT optimization.
  7. Allistah


    Thanks for the info. This was helpful. I sent you a private message as well.
  8. Quants can be a little strange......

    Tradestation 9.0 backtests have worked well for me. New equity highs for this TA based aglo that I made over a year ago; 9 lines of eazylanguage:
  9. Allistah


    Funny you say that because I just moved over to Tradestation yesterday.

    I'm working on making some strategies work and getting used to the tools they have for back/forward testing.

    How do you like the walk forward optimizer? Do you have any "Make sure you..." type of tips for me on Tradestation? I have $2.36 commissions being added per contract.
  10. I love Tradestation and will probably never leave, unless they change in some major way.

    I played with the WFO tools last year to confirm that I had not made any major mistakes, and my backtests didn't change much, so it came as a good sign. I also had Portfolio Maestro for about a month and that confirmed some other ideas (I started a thread about it somewhere, you may find it entertaining).

    One thing I did find to be extremely important is the overall trend of the stock, etf, or futures contract (for long - only algos).

    The attached algo is long & short so it may be immune, but I had a couple small cap algos break down when the stock's personality changed after the back test was finished.

    Another surprise was how different SPY was from ES..... totally different solutions were found due to the different price action

    #10     May 4, 2013