Best
Tim

Would it be possible to implement such feature into FSB as well?

Best regards

We have that update in our ToDo list. Please stay around and you will hear for the update when it is ready.

it looks very promising. It gives another way of optimization.

I wonder if to include percent at gain can be helpful

#    Before Optimization    After Optimization    Gain
1    2769                                   2120                     -649   (23%)
2    2270                                  1952                     -318   (14%)
3    2219                               2007                      -212 (9.5%)

thanks in advanced! cheers from Brazil

Fortunately, you can solve this problem with EA Studio:
- Decide what time span you will use for creating your strategy and for testing on unseen data ( OOS ).
- Set the first time span in Data Horizon and built and test the strategy.
- Set the second time stamp in the Data Horizon (You may overlap the data span). EA Studio will show you the full performance of the strategy on the OOS data.   
- Use Walk Forward to see if the strategy will perform better, if you optimize it on segments.

Looks very good Mr. Popov, good work!

So if at all, this should become and OPTION, not something that is forced upon the user.

Thanks

Excellent point @geektrader... Totally agree that "unseen data for walk forward" should be an option, not a static feature :-)

@moreirajbn: I understand what you say, but different users = different approaches. I use it for parameter stability testing, so data snooping (using the same data the strategy was generated on) is not an issue in that case and I do not like to be forced data that has "not been used before", especially since I create my strategies on 32 years of data without any OOS. So if at all, this should become and OPTION, not something that is forced upon the user.

Thanks

Actually I will spend the next two days in making the Walk Forward more powerful.

Since you're working on improving WFA, I have a suggestion:

As it is now, it's running the walk forward process on market data that the strategy development process has already been exposed to (the same period from where the strategy was discovered). So, technically the WFA is suffering from data snooping, because it's "walking" on the same data used to develop the strategy. For example: if 2005-2010 is used in the "Data Horizon", the Generator is looking for strategies that work on 2005-2010 period, and then the Reactor runs WF validation on the same period.

To simulate the "true" walk forward concept, the in-sample period used to discover and optimize the strategy should be completely independent of the out-of-sample, since this is what will happen when we are developing a strategy to run in the future (we have data from XXX to June-2018, and the key challenge is to be profitable after that).

So, my suggestion is: if the user selects walk forward validation in Reactor, and 2005-2010 is selected in "Data Horizon", only the first in-sample period should be available to Generator. Let's say that in-sample period is 1 year, then the data mining process should only run from Jan-2005 to Dec-2005, and the remaining period (2006-2010) used for walk forward validation.

Awesome work you're doing here!!!

I'll add chart line of the initial equity before the optimization of each segment.

I'll add also tables for all stats params like that:

Net Balance
Segment #, Original , Before Optimization, After Optimization, Gain

Later I'll think of implementing Matrix calculations. My concern about it is that the output may become blotted with stats data.

It has the Walk Forward segments calculated correctly and also some other minor fixes.

Please reload the app to use the newest version.