Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Jung-eun Lee <tinymoon@yahoo.com> |
To | STATA <statalist@hsphsun2.harvard.edu> |
Subject | st: Residuals after count model (nbreg) |
Date | Tue, 11 Oct 2011 11:14:09 -0700 (PDT) |
Dear STATA list users, could anybody help me find the residuals from my analysis (negative binomial regression) and test if they are correlated with time? I am using event count modeling. Is this a correct method?: "predict res" or "predict dev" -- to calculate residuals? The results seem to be the same. Then what do I do? Should I create a plot with res(or dev)(y axis) and time(x axis)?("twoway (scatter res date)") What should I look for in the graph?I will greatly appreciate your help. * * 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/