Re:[R] exponential smoothing

From: Adrian Trapletti (adrian.trapletti@lmttrading.com)
Date: Mon 17 Jun 2002 - 17:53:58 EST

• Next message: Christian Hennig: "[R] OT: Journal of Statistical Software"
• Previous message: Peter Dalgaard BSA: "Re: [R] lm() function to get coefficients doesn`t work"
• Maybe in reply to: Erin Hodgess: "[R] exponential smoothing"

Message-id: <3D0D9596.88FB16A4@lmttrading.com>

> Date: Fri, 14 Jun 2002 13:21:23 +0200
> From: Xavier.Abulker@fimat.com
> Subject: [R] exponential smoothing
>
> could someone help me to write a fonction doing an exponential smoothing in
> case of a multivariate time serie?
> I tried
> ewma <- function (x, lambda = 1, init = 0)
> {
> if (is.ts(x))
> filter(lambda*x, filter=1-lambda, method="recursive", init=init)
> else
> stop(message="first argument should be a time serie")
> }
> but I can't apply that to multivariate
> Thanks
>
> xavier
>

For example:

apply(x,FUN=ewma,2,lambda=0.1)

should work. You maybe need to set the default of init in ewma() to x[1].

best
Adrian

--
Dr. Adrian Trapletti   Phone:       +41 (0)1 994 56 31
Wildsbergstrasse 31    Fax  :       +41 (0)1 994 56 33
CH-8610 Uster          Email:   a.trapletti@bluewin.ch
Switzerland
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

• Next message: Christian Hennig: "[R] OT: Journal of Statistical Software"
• Previous message: Peter Dalgaard BSA: "Re: [R] lm() function to get coefficients doesn`t work"
• Maybe in reply to: Erin Hodgess: "[R] exponential smoothing"

This archive was generated by hypermail 2.1.3 : Wed 16 Oct 2002 - 11:57:29 EST