0

I'm currently trying to build a model with one predictor variable. I have insurance data: loss dollars and their policy limits.

I'm able to plot a histogram with the loss amounts and fit a gamma distribution to the data, but I don't get where the limit factors in. The only model I've built with limits and loss is a simple linear y = mx + b model.

This is my first attempt at modeling, if I'm missing something fundamental or basic, please let me know.

Player2Entered_
  • 1
  • Does this answer your question? [How would you fit a gamma distribution to a data in R?](https://stackoverflow.com/questions/45536234/how-would-you-fit-a-gamma-distribution-to-a-data-in-r) – csgroen Aug 03 '20 at 08:19
  • Are you trying to fit a conditional (on policy limits) gamma distribution? If so, use a Gamma GLM. – Roland Aug 03 '20 at 09:13
  • I dont think the first link answers the question since it doesn't take in a predictor variable, it just fits gamma to the observations/response variables. – Player2Entered_ Aug 05 '20 at 04:39
  • Gamma GLM could work! I'll look into it some more. I initially had a very simple linear model, but if I can convert it to gamma that would be great. – Player2Entered_ Aug 05 '20 at 04:40

0 Answers0