[R] Rolling Correlations

Gabor Grothendieck ggrothendieck at gmail.com
Tue Dec 4 09:31:19 CET 2007


If you want to use just zoo alone then its done like this.  See
?rollapply and the two zoo vignettes.

> library(zoo)
> set.seed(1)
> z <- zoo(cbind(a = rnorm(25), b = 1:25))
> rollapply(z, 5, function(x) cor(x[,1], x[,2]), by.column = FALSE)
          3           4           5           6           7           8
 0.54680491 -0.13300909  0.03576759 -0.28297639  0.52062842  0.26441877
          9          10          11          12          13          14
 0.24494114  0.05774863 -0.32294217 -0.68658780 -0.35636555  0.11012008
         15          16          17          18          19          20
 0.43888843  0.61623274  0.10862581  0.71947286  0.60338376 -0.25653111
         21          22          23
-0.61338395 -0.79316968 -0.44360374


On Nov 30, 2007 7:38 AM, Shubha Vishwanath Karanth
<shubhak at ambaresearch.com> wrote:
> Hi R,
>
>
>
> I want to do some rolling correlations. But before, I searched for
> "?rollingCorrelation" and tried the example in it. But I was not
> successful. What could be the problem? Here is the code I tried:
>
>
>
> > library(zoo)
>
> > library(PerformanceAnalytics)
>
> > rollingCorrelation(manager.ts at Data[,1],edhec.ts at Data,n=12)
>
> Error in inherits(object, "zoo") : object "manager.ts" not found
>
>
>
>
>
> BR, Shubha
>
>
>        [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>



More information about the R-help mailing list