Hi, I tried with lars, but for a 29(n)x4662(p) dataset, I kept getting
error. One is the program failed,giving out a message:"if (zmin <
gamhat)
{missing value where TRUE/FALSE needed", another problem, for leave one out
cross validation with cv.lars, the program sometimes gives very large
ridiculous prediction on omit sample(s). Anyone has similar problems?
Thanks.
[[alternative HTML version deleted]]