I suggest you to consult this books Sheldon M Ross :

ROSS, Sheldon M. Introduction to probability models. Academic press, 2014.

ROSS, Sheldon M. Introduction to probability and statistics for engineers and scientists. Academic Press, 2014.

Thanks

If you have variable  which has negative binomial distribution than it is usual that standard deviation is higher then mean.

You should estimate parameters of negative binomial distribution r, p (using annotation from https://en.wikipedia.org/wiki/Negative_binomial_distribution) using (Y is a variable with negative binomial distribution) and analyse variability of these parameters  r and p 

proc genmod;

model Y = / dist=negbin;

ods output parameterestimates=pe;

run;

Stata examines this easily when comparing  the Poisson to the Negative binomial model :-)

Hi Sasan, I was wondering if you finally found a way to calculate the overdispersion parameter in negative binomial distribution, because i now have the same question to calculate a sample size for a clinical trial.

I only found some information about it and is something like doing simulations but it is not detailed!

Dear professional contributors,

I need help in providing model parameters derivations for Joint Modeling of Multivariate Repeatedly Measured count Outcomes. email: [email protected]

Thank you so much.