No announcement yet.

EFS... The most difficult trade script!

  • Filter
  • Time
  • Show
Clear All
new posts

  • EFS... The most difficult trade script!

    How many of you saw anyone write a public trade script using EFS? Search on Google, it is not only "few", but it is absolute "rare".

    I've use eSignal for years. EFS study is always a pain to develop. I've read all the available material I can find in eSignal guide. However, these material are so outdated which date back to eSignal 10. Many functions in eSignal 12 are missing, incomplete or useless. The EFS KB is not available for years. So there is hardly anything I can refer to.

    I don't know why eSignal make EFS so complicated and confusing. For a trade script, it should be easy to use and simple to understand within hours. Base on Javascript, EFS "should be" very powerful with OO capability. However, for an OO language, you need a good debugger to work with it. All we have is a debugPrintln(). That's all! Nothing else. The worst thing is, some studies come with eSignal, it is 100% working. But when you try to trace its variable changes during execution, its value is always NaN. So I got a working study but cannot trace its value change?

    There is 2 kind of input variables. You can use simple VAR xxx or VAR new Array(). So what's the difference between these 2? All I know is it seemed you cannot debug a simple var variable during runtime. So if the developer does not know how the var value changes, how can he debug his study? An OO is very powerful! But how many of you ever use EFS OO capability?

    If eSignal like to make EFS powerful, please do like what MetaTrader do with their EA. If eSignal wants to make EFS popular, please do like ThinkOrSwim script. Just a plain brain dead script will be sufficient for most traders. A simple trade script should be done in several lines, not in hundreds of line of code to show how good you are.

  • #2

    As a user of eSignal for many years I agree with many of your points. It is unfortunate for their customers that unlike other vendors you will not find a single functioning sample EFS trading strategy that will in real time execute trades using the eSignal Paper Broker using current generic broker functions.

    I published the only one I've ever seen on the forum, that at the time in 2008 when the volatility was extraordinary was actually very profitable

    You should however be able to trace any variable using debugPrintln() I trace each variable in my code and do not get that error can you post an example?

    It's safe to assume most users are not looking at charts for academic reasons but to hopefully trade profitably. One would think with the millions of dollars customers have paid eSignal that they would provide a workable trading template that can be enhanced or modified with some potential to be used as a profitable strategy, as many other vendor's provide.

    Choosing not to publish examples that would help customers utilize the product as is standard with other vendors is one issue as you indicated.

    A much more fundamental problem is the back tester in particular, over inflates the profitability of the most basic trading strategy when multiple time intervals, a feature that is a part of the product for many years now.

    As support requested when the incident was opened I documented and posted this issue to the forum clearly demonstrating the hyper inflated profit result reported by the back tester many months ago, and have not heard back regarding the back testing problem.

    The strategy utilizes multiple time intervals and the profits reported when back tested compared to when executed real time are drastically different, going well beyond any slippage, a clear programming defect or "bug".

    I'm not sure if the eSignal community at large is interested in the product functioning in-line with the advertised features but if so by others not weighing in that will unfortunately never happen

    As bad as eSignal's lack of support would be for even the smallest of firms. The fact that this is an ICE/NYSE subsidiary selling a product perhaps unintentionally, with grossly exaggerated profits could appear that they are in effect fleecing their customers is astonishing. One would think they would mitigate risk t by having a developer resolve the issue, or explain the reason for the hyper inflated results in my posts.

    Having used the product for years and familiar with the support staff I know this is not their intent but can easily be misconstrued to be the case.

    Which is why I am hopeful that they simply address this one individual issue with the back tester.

    Hope this helps.
    Last edited by demarcog; 08-21-2019, 08:30 AM.
    Glen Demarco