Re: [R] Composite Quantile Regression

This reply is more than a little late, but I just ran across Frank's question today as I was searching for something else in R-help. And as it happens I was about to incorporate some old code to do "Composite QR" into the quantreg package. This new version has now been uploaded to CRAN so, with so

Re: [R] Change Variable Labels in Quantile Plot

Hi Roger, Maybe I'm missing a clue. Here's an example: x<-rnorm(50) x1<-rnorm(50) y<-x+x1+rnorm(50) out<-rq(y~x+x1, tau=1:9/10) plot(summary(out)) plot.out<-plot(summary(out)) #I change the variable names dimnames(plot.out)<-list(c("intercept", "sex", "inc")) Why not just change the names

Re: [R] Question of "Quantile Regression for Longitudinal Data"

I've used the function rq.fit.sfn and rq.fit.panel to estimate a quantile regression on a panel data set.Now I would like to compute an statistic to measure the goodness of fit of this model (T-statistics and P-value ). Does someone know how could I do that? For formal inference you are be

Re: [R] lty dots pdf issue

Thanks Ken, that was it. I've updated to R version 2.11.0 Under development (unstable) (2010-02-18 r51149) x86_64-apple-darwin9.8.0 and everything looks fine now. Thanks too to Simon for sorting this out! Roger -- View this message in context: http://n4.nabble.com/lty-dots-pdf-issue-tp1561