On Tuesday 14 October 2003 17:41, Laura Quinn wrote: > As a total beginner still, I am having great difficulty trying to > apply ar models to column data in a matrix. > > As each matrix has 20 columns (and some of these contain NA data) it > is obviously a long, slow process to calculate ar models for each > column individually, though I am (probably extremely stupidly) > struggling to see how I am doing this wrong. > > I have tried using: > > for (i in 1:20) > apply(mydata,c(,i),ar(mydata[!is.na(mydata)],order.max=3)) > > and > > apply(mydata,ar(mydata[!is.na(mydata)],order.max=3)) > > and > > apply(mydata,c(,1:20),ar(mydata[,1:20][!is.na(mydata[,1:20])],order. >max=3)) > > and none of these work - can someone tell me what I am doing wrong??
Try apply(mydata, 2, function(x) ar(x[!is.na(x)])) and then look at help(apply) to understand what you have done wrong. hth, Z > Thank you in advance, and excuse me for my stupidity.. > Laura > > ______________________________________________ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help ______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
