So if you have 10 systems, and you want to create a portfolio of the top 5 least correlated systems, how would you calculate the correlation?
lets say they are all on different timeframes and some might wait 1 day or 1 hour or 1 week before doing its next trade so obviously dates would have to be involved somehow.
We want to test the correlation of the data series, so if you have 3 systems that trades almost always at the same time, they are surely correlated somehow.. So in order to get the most from the correlation analysis, we should add time into the equity data series as the primary coordinate and not with trade numbers. Hope i’m clear 🙂
Hi Nicolas, yes this is what im thinking as well, im just not sure how to treat systems where 1 system might have 1000 trades and another might have 500 in the same amount of time.
Should i for example use “correlation per weekly gain/loss”?
To help us continually offer you the best experience on ProRealCode, we use cookies. By clicking on "Continue" you are agreeing to our use of them. You can also check our "privacy policy" page for more information.Continue