Calculating trailing returns

On Mon, Oct 26, 2015 at 9:53 AM, Am Gut <agquantr at gmail.com> wrote:

Good Morning Michael,

I have simple return data, in a daily periodicity. I am essentially
trying to calculate the trailing returns for say 252 periods, assuming I am
trying to look at trailing 12 month returns. So I have been trying to use a
product function, but I am having trouble specifying a look back period. I
did use the apply.rolling function from the PerformanceAnalytics package,
but was unable to us the prod function with it- the function was simply
returning an average:

You're not passing the function argument correctly (you're passing
'trim=function(x) prod(1+x)-1'). Check the function signature.

library(PerformanceAnalytics)
data(managers)
M1 <- managers[, 1, drop=FALSE]

apply.rolling(M1, width=5, FUN=function(x) prod(1+x)-1)[5] ## cumulative
return of first 5 observations

## compare to manual calculation
prod(M1[1:5]+1) - 1


Michael

##compute 12-month (lookback variable) returnm
List = list()
for (i in 1:ncol(sector_xts))
{
  List[[i]] = apply.rolling(sector_xts[,i], function(x) prod(1+x)-1,
width = 252)

}
lookback_returns = do.call(cbind,List)