Goodday to all,
I want to compute the Dunn & Smyth's randomized quantile residuals for a
gamma model estimated using the glm-command.
After reading the paper I have some doubts as to how the residuals are
defined. It is really just the standard normal quantile function of the
fitted value?
That is:
predict ybar, xb
invnormal(ybar)
Just want to check whether this is correct.
Thanks for your help!
Serguei Kaniovski
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
