Last active
December 24, 2024 07:06
-
-
Save dadaromeo/33e581d9e3bcbad83531b4a91a87509f to your computer and use it in GitHub Desktop.
Implementation of the Conway-Maxwell-Poisson distribution in pymc3
+1 to adding this to PyMC3
Hi, many thanks for the job done!
When i try to use it with nu less than 1 the code loop forever in the cycle "while any(u > cdf):"
You can try with "CMPoisson.dist(lamda= 1.24598558, nu=0.5681564).random(size=1000)"
I do not have the right mathematical background to solve the problem.
Do you have any guess or direction ?
Thanks a lot!
Regards.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi! May be you would like to add this distribution to PyMC3! I interested about learning how this distribution compares with the negative binomial one (that is also used to model over-dispersed data). I was unable to open the first reference it seems that the correct link is www.galitshmueli.com/system/files/JRSS-COM-Poisson.pdf).