# Programming a mixture of a Gamma with a Normal distribution using R

I have some data x in R which seems to be a mixture of a Gamma and Normal distribution. Therefore I'd like to model this as a mixture model consisting of said distributions, but I don't know how to fit this in R. I'd prefer to estimate the model using maximum-likelihood or the EM-algorithm, but something else would be nice too. There are numerous packages for doing this for mixtures consisting out of only Gaussians or Poissons etc. But I can't find anything for mixed models consisting of more than one distribution family, and I'm also kind of a novice to R so I wouldn't know how to implement this myself.

## migrated from stackoverflow.comMar 9 '17 at 2:02

This question came from our site for professional and enthusiast programmers.

• Yes DWin but focusing on programming is off topic for CrossValidated too. – Michael Chernick Mar 9 '17 at 2:46
• Perhaps the questioner will be satisfied with: stackoverflow.com/questions/15823320/… . I thought the question deserved a theoretic discussion, but perhaps he wanted something that just gives an answer, albeit an answer with little analytic underpinning (with no offense intended to Ben Bolker). The other thought (if one were constrained to just the R language) might be to offer a simple R-focussed search strategy: sos::findFn("mixture model gaussian gamma"), after which I see at least one candidate. – DWin Mar 9 '17 at 17:43