Dear all,
I would like to explain a quantitative variable with a very skewed
distribution using some explanatory variables. Obviously linear models
are not appropriate because of the skewness of the dependant variable
distribution. I thought of using generalized linear modelling with a
gamma distribution, but I'm not sure whether it is appropriate. What do
you suggest? In particular, I wonder how to check the validity of the
gamma distribution.
Thanks a lot in advance.
Regards,
Eve
|