R help - Jan 2010 - problems about Quantile Regression for Longitudinal Data

Hello,

I uesd Dr. Koenker's code on the following website for quantile regression
of longitudinal data:

http://www.econ.uiuc.edu/~roger/research/panel/rq.fit.panel.R

I can run the code when there are three explanatory variables in X matrix
(ncol(X)=4, because there is a vector "1" at the first colume).
However, when I increase the number of explanatory variables to four or more
than four, R showed me an error message "local(x, pivot, ...) : Increase
tmpmax."

Could I please ask, how do I deal with this problems?
By the way, I set w=c(.1,.2,.3,.4,.5,.4,.3,.2,.1),taus=(1:9)/10, and lambda=1.

Many thanks for help.

--
Chu, Hao-pang, Graduate
Department of Economics
National Chengchi University
e-mail:97258031 at nccu.edu.tw