« Fitting mixed models to categorical data in R | Main | A tutorial on Support Vector Regression »

May 14, 2009

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a010534b1db25970b0115707fb47e970b

Listed below are links to weblogs that reference Parallelized Backtesting with foreach:

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

David (and Bryan),

This is awesome! Thanks for sharing the code!

One (trivial?) comment:
I was a bit confused about where the INTC and IEF data came from until I got to the end of the post. I would have expected to see a "load('Vhayu.Rdata')" command before the chartSeries call.

great post, but is there not a lagging issue here? don't you need to lag the signal by a day to make sure you are not capturing the return on the day that the macd signals (i.e. removing a current day look forward?)

there's also an error in the current code.

The line

return (long*(s>0) + benchmark*(s<=0))

should be

return(rbind(long[s > 0],benchmark[s <= 0]))

@Brian, I *think* the code is correct. (Going by memory here -- I haven't rerun the code lately.) It's unusual R syntax, but it's returning either the value "long" or "benchmark" according to the value of "s" -- treated as numeric, (s>0) is a sequence of 1's and 0's.

I completely understand what you were getting at, and I remember running this example when you first posted it.

I needed to make the indicated modification to get it to run on now-current versions of everything. Hopefully it will help someone searching old posts.

Regards, Brian

can this be run on a windows machine

Hello and thank you for this post. I have the following problem maybe you could help me out with it?

1. I get this error which seems to be telling me that the "simpleRule" which defines the signal and orders it using "z" needs an appropriate time-based object. SEE BELOW

#### CODE HERE
R <- simpleRule(z$INTC, fast=12, slow=26, signal=9, long=Ra, benchmark=Rb)
#### ERROR RETURNED
Error in xts(position, order.by = index(z)) :
order.by requires an appropriate time-based object

I have no idea why I'm getting this error. If I check the str(z) it is an xts object.

Thank you,
Douglas

It looks as if can skip the s <- xts(position,order.by = index(z))
I made it a comment by adding # before it.
Then when applying the rule I past the "position" into the rule! I maybe missing something but the position is and xts object and needs not be converted?
return(long*(position>0)+benchmark*(position<=0))
Below is the code and how I changed it. Any feedback would be great! Also, I think you have a lag issue? While I'm asking questions, what would be the quickest way to subset the data to test different time frames?
Thank you,
Douglas

simpleRule <- function(z, fast=12, slow=26, signal=9, long, benchmark)
{
x <- MACD(z, nFast=fast, nSlow=slow, nSig=signal, maType="EMA")
position <- sign(x[,1]-x[,-2])
#s <- xts(position,order.by = index(z))
return(long*(position>0)+benchmark*(position<=0))
}

#####USE THE RULE NOW
R <- simpleRule(z$INTC, fast=12, slow=26, signal=9, long=Ra, benchmark=Rb)
chart.CumReturns(R, main="Nfast=12 NSlow=26 NSig=9")

The comments to this entry are closed.


R for the Enterprise

Got comments or suggestions for the blog editor?
Email David Smith.
Follow revodavid on Twitter Follow David on Twitter: @revodavid

Search Revolutions Blog