[R-SIG-Finance] Calculating historical returns of a set of trade rules over multiple assets

Charlie Friedemann cfriedem at gmail.com
Mon Sep 5 05:47:44 CEST 2011


Greetings,

Recently, I've been pushing more of my work into R. However, I feel  
like I'm missing out on some of the faster methods of doing things,  
which is why I turn to the people on this mailing list for help.

Anyway, one thing I'd like to use R to do is to calculate the trailing  
returns of a set of about 60 trading rules (many are the same rule  
parameterized over various window lengths) for a list of about 250  
asset pairs, such as SPY-IWM.

I know it's a pretty general question, but is there any way you guys  
would approach doing this in R that would be a generalized, almost  
vectorized, approach? I'd rather not have to do a for loop type  
approach over each pair and each indicator, as I feel like that  
wouldn't be as fast as doing them en masse (if that's even possible).  
This is part of an analysis that I need to run daily so speed is  
definitely an issue.

Oh, and I'm on Windows (Linux is not possible as others on the team I  
work with can't/won't/don't use it). I know this is a very general  
question, but even just some pointers as to what packages/approaches  
to check out would be very useful. I'm sure others on this list have  
had to deal with something similar.

Thanks!



More information about the R-SIG-Finance mailing list