I am trying to fit a model with a glm using a tweedie family. I use a index parameter p between 1 and 2 to get a compound Poisson Gamma distribution to fit my data. But I want to use an offset only on the Poisson part of the regression. So my questions are the following ones:

  • Can this be done and how?

  • Is it possible to split the two part of the regression (Poisson and Gamma) manually to put the offset only on the Poisson part?

  • What do you known about the way that a tweedie regression handles an offset?

  • $\begingroup$ An offset in GLMs is a shift to the linear predictor; if you use a log-scaled variable as an offset in a model with a log-link you would get a multiplicative effect on the scale of the response (suitable for adjusting for exposures, say). This will be the same for Tweedie as anything else. $\endgroup$ – Glen_b -Reinstate Monica Sep 3 '17 at 9:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.