Not logged in

Forums

Viewing 15 posts - 1 through 15 (of 19 total)
  • #113375 |
    Customer
    90 Posts

    Mark,

     

    What happened to Quant Strategy EA??

    #128992
    Administrator
    3055 Posts

    unfortunately, the results were not very good recently, so we stopped offering it. 

     

    I’m also too busy with programming as we are focusing on new SQ 4 and have very little time to develop actual trading strategies right now.

    Mark
    StrategyQuant architect

    #129033
    Customer
    312 Posts

    Results not very good?

    Strategy was curvefitted? To be honest, Mark,if your flag strategy generated on flag software fails, then its not good for whole product.

    btn_viewmy_160x33.png

    #129045
    Customer
    905 Posts

    The more I look into testing and curve fitting I’m coming to the conclusion there are only two types of automated strategies. Curve fitted strategies that happen to work for a while and curve fitted strategies that are no longer working.

     

    Hope I’m wrong though.

    #129052
    Administrator
    3055 Posts

    curve fitting is the biggest problem for strategies created using any machine learning approach, but it doesn’t mean that all automatic strategies are curve fitted.

     

    The problem arises also from to much improvements – the more you try to optimize or improve the strategy to perform well on history data the more you curve fit it.

     

    Adrian – I understand your point, the problem is that because of programming I have no time to develop strategies.

    I think it is better to stop using / selling something that I’m unable to maintain and support.

     

    If you are looking for proof that SQ can generate good strategies we’ll publish an article with live results soon .

    Mark
    StrategyQuant architect

    #129080
    Customer
    312 Posts

    Adrian – I understand your point, the problem is that because of programming I have no time to develop strategies.

    I think it is better to stop using / selling something that I’m unable to maintain and support.

     

    If you are looking for proof that SQ can generate good strategies we’ll publish an article with live results soon .

     

    I’m waiting for such article ! 

     

    And if it’s about no time for develop – Mark, to be honest, I think if develop would give You 100 or 200% profit annually, then surley You would give up with programming and start do developing 🙂 But…. That’s only my point of view. 

    btn_viewmy_160x33.png

    #129093
    Administrator
    3055 Posts

    And if it’s about no time for develop – Mark, to be honest, I think if develop would give You 100 or 200% profit annually, then surley You would give up with programming and start do developing  🙂 But…. That’s only my point of view. 

     

    you are right, but not quite.

    I’m good programmer, but I’m not sure if I’m a good trader. I was always interested more in the programming and technical side of things than in actual trading.

    It is perfectly possible to develop a great portfolio of strategies even with current or older versions of SQ, as some of my customers did, but I’m still working fulltime on the new version,

    instead of spending my time developing my portfolio of strategies.

     

    Since I started StrategyQuant I really haven’t had time to “only” develop strategies, also because developing and maintaining  product like StrategyQuant is a fulltime job.

    It is not only about having the tools, but also the right mindset, correct work processes, and time in your hands.

    Mark
    StrategyQuant architect

    #129098
    Customer
    312 Posts

    It is perfectly possible to develop a great portfolio ,as some of my customers did, 

     

    I wish to know them 🙂 Do they dissapeared from forum? 

     

     

     

     

    I’m good programmer, but I’m not sure if I’m a good trader.

     

    SQ isn’t made to be a good trader. It is tool that creates automated strategies . There is no need to be a good trader or even know trading methods – beside those basics.

     

     

    Anyway, You have good money from sales – I think even more than from trading at small capital 😉

    btn_viewmy_160x33.png

    #129227
    Customer
    80 Posts

    I hope nobody is under the illusion that it’s possible to make a strategy that is going to be profitable 100% of the time? There is enough randomness in any market to make this task impossible. The only thing we can hope for is that the profits, over time, are larger than the drawdowns.

     

    In order to accomplish that, it may be necessary to run a portfolio of none-correlated strategies even on the same FX symbol. The new  EA Analyzer can do just that, make a none-correlated portfolio of strategies… 

    #129325
    Customer
    57 Posts
    I’ll say it that simple. SQ works! But it is you who has to make it work.
     
    Mark, I’m out the forum long ago. But I come to give you thanks and congratulations for your work
     
    Best regards

    AndrEAs

    #129339
    Customer
    481 Posts

    Hi,

    yes its works I think. We can observe the myfxbook account the nexth months.

     

    Importent is the way of generation and filtering:

    I think no one will tell all other the way of finding if its works. I am searching 3 years for a good way.

    I learn every day a little more

    It will give me more energy to find the solution if someone has found a solution 🙂

     

    thomas

    #129390
    Customer
    598 Posts

    SQ works for me – I think everyone defines what “works” means differently.

     

    For me it´s the first product that gives me ALL the possibilities I was ever looking for for generating automated strategies. Do the strategies work trading live? Yes and no, just like with any strategy that was ever developed and that still works or doesn´t work anymore. Automated OR manual, it doesn´t really matter – any strategy can fail any day and you´ll NEVER EVER know when BEFORE it happens.

     

    I am doing automated forex trading full time for the last 8 years now and I am NOT rich, but I can make a steady income from automated strategies. Sure sometimes you can get into drawdown or side-ways with your income, but the main point is to not lose to much money (surprise eh?;)).

     

    My advise and way of doing it is this: generate the strategies, see that they have enough trades so that you have a valid stasticial edge at lest for the backtests (for 2001 to date I only use strategies that had AT LEAST 800 trades) don´t play TO much with robustness and walk-forward etc. (just my experience as such tests never helped me anything going forward live, it made no difference!). You anyway get a feel for what works the longer you practice. Then create many of such strategies for different pairs / timeframes – create a huge portfolio with low correlating strategies (SQ displays that, so that´s an easy one)! Then put the portfolio live and over time disable the strategies that do not work anymore (for me the time to disable them is when they have exceeded their max historical backtest drawdown) with new ones you´ve discovered meanwhile. For me it´s a daily ongoing process and SQ has made it a 1000 times easier for me than it ever was before when I had to code every strategy in MQ4 and test it with the crappy and slow strategy tester.

     

    So yes – SQ works for me!

    #134283
    Customer
    362 Posts

    SQ works for me – I think everyone defines what “works” means differently.

     

    For me it´s the first product that gives me ALL the possibilities I was ever looking for for generating automated strategies. Do the strategies work trading live? Yes and no, just like with any strategy that was ever developed and that still works or doesn´t work anymore. Automated OR manual, it doesn´t really matter – any strategy can fail any day and you´ll NEVER EVER know when BEFORE it happens.

     

    I am doing automated forex trading full time for the last 8 years now and I am NOT rich, but I can make a steady income from automated strategies. Sure sometimes you can get into drawdown or side-ways with your income, but the main point is to not lose to much money (surprise eh? ;)).

     

    My advise and way of doing it is this: generate the strategies, see that they have enough trades so that you have a valid stasticial edge at lest for the backtests (for 2001 to date I only use strategies that had AT LEAST 800 trades) don´t play TO much with robustness and walk-forward etc. (just my experience as such tests never helped me anything going forward live, it made no difference!). You anyway get a feel for what works the longer you practice. Then create many of such strategies for different pairs / timeframes – create a huge portfolio with low correlating strategies (SQ displays that, so that´s an easy one)! Then put the portfolio live and over time disable the strategies that do not work anymore (for me the time to disable them is when they have exceeded their max historical backtest drawdown) with new ones you´ve discovered meanwhile. For me it´s a daily ongoing process and SQ has made it a 1000 times easier for me than it ever was before when I had to code every strategy in MQ4 and test it with the crappy and slow strategy tester.

     

    So yes – SQ works for me!

     

    SQ works for me TOO!

    #134311
    Customer
    48 Posts

    SQ works for me TOO!

     

    Since I started my EA life in October ((EA generated by SQ), I am making an average profit of $40/ day

     

    Now I am continue to use SQ to generate variant of EA to increase my portfolio, to spread the risk 

     

    So far so good and I am glad I made the investment, couldn’t wait for SQ 4

     

     

    M.K  

Viewing 15 posts - 1 through 15 (of 19 total)

You must be logged in to reply to this topic.