Discussing the article: "Combinatorially Symmetric Cross Validation In MQL5"
MetaQuotes:
Check out the new article: Combinatorially Symmetric Cross Validation In MQL5.
Author: Francis Dube
Just curious if anyone has had luck with this method? i tried implementing it on an m5 backtest over 10 years with a forward at 1/2 and it's insanely slow, would like to know if anyone found a way to code it so it's a little faster?? sure would be interesting to try this method out though.
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Check out the new article: Combinatorially Symmetric Cross Validation In MQL5.
In this article we present the implementation of Combinatorially Symmetric Cross Validation in pure MQL5, to measure the degree to which a overfitting may occure after optimizing a strategy using the slow complete algorithm of the Strategy Tester.
Sometimes when creating an automated strategy, we start out with an outline of rules based on arbitrary indicators, that need to be refined in some way. This process of refinement involves running multiple tests with different parameter values of the chosen indicators. By doing so we are able to find the indicator values that maximize profit or any other metric we care about. The problem with this practice is that we introduce a certain amount of optimistic bias because of the prevalent noise in financial time series. A phenomenon known as overfitting.
Whilst overfitting is something that cannot be avoided, the extent to which it manifests can vary from one strategy to another. It would therefore be helpful to be able to determine the degree to which it has occured. Combinatorially Symmetrical Cross Validation (CSCV) is a method presented in an academic paper "The Probability of Backtest Overfitting", written by David H. Bailey et al. It can be used to estimate the extent of overfitting when optimizing parameters of a strategy.
In this article we will demonstrate the implementation of CSCV in MQL5 and show how it can be applied to an Expert Advisor (EA) through an example.
Author: Francis Dube