I’ve just been through my first workflow on the builder. I went from 1500 to 3 strategies and whilst they were succesful in the robustsness testing they failed the last ou of sample test on new data.
I was following the 2011-2016 on main sample data. I then used 2003- 2010 for the first out of sample. Then I ran 2017-2021 on the last test. 3 years was a long time compared to the 1 year sample the video showed. They all failed with small losses.
I am running a new optimisation using 3000 strategies and am wondering- should I move the main sample to 2015-2020 leaving the 2021 as one year for final OOS testing?
Or perhaps 2011-2019 for the first sample? Using 2003-2010 old data for first OOS tests?
2 weeks ago #274434
you will find it difficult to create strats which work over the span you have chosen. There have been many market regimes.
only 2cents: beware that it seems, that you are using UTC0 original dukascopy data and many brokers are UTC2 timezone, so the candles are looking different
you need to clone your data to a proper timezone of your broker
You want to be a profitable algotrader? We started using StrateQuant software in early 2016. For now we have a very big knowhow for building EAs for every possible types of markets. We share this knowhow, apps, tools and also all final strategies with real traders. If you want to join us, fill in the FORM. 1500+ final SQX strategies for members running on demo accounts to verify the edge DEMO ACCS. We provide also strategies for indices - DAX, DOW JONES, NASDAQ, SP, UK, because we have realtick data from our brokers.
Thank you I appreciate that. I’ll dig up how to do that unless you can point me to any guides I might have missed?
Viewing 4 posts - 1 through 4 (of 4 total)