[R] glm with Student t for error distribution

Jonas Malmros jonas.malmros at gmail.com
Fri Oct 26 10:23:16 CEST 2007


My response variable seems to be distributed according to Student t
with df=4. I have 320 observations and about 20 variables.
I am wondering whether there is a way to fit glm with Student t for
error distribution. Student t is not one of the family choices in glm

How should I proceed to fit glm with Student t?
I know that Student t is the Inverse Gamma with shape parameter equal
to degrees of freedom (=4). Would it be correct then to specify Gamma
family and inverse link in the glm function?

Thank you for your help.

Jonas M.
University of Stockholm

Jonas Malmros
Stockholm University
Stockholm, Sweden

More information about the R-help mailing list