We are trying to obtain a predictive distribution in a Poisson model using a non-informative prior distribution. We have found a gamma poisson as a result of our analysis and we want to evaluate it in R. Is there any package where this distribution and anothers are defined? In any case, we have tried to programm the gammapoisson probability function but we have found another problem: we must evaluate a gamma function with a quite high value (gamma(284)). How can it be done? Thank you very much!
Bayesian package
2 messages · Ramon Martínez Coscollà, Spencer Graves
2 days later
GAMMA-POISSON: Am I correct that the discrete marginal of a gamma-Poisson is negative binomial? Try, "?pnbinom" for that. The posterior is another gamma; use "?pgamma", etc., for that. GAMMA FUNCTION WITH LARGE ARGUMENTS: > lgamma(284) [1] 1318.403 > gamma(284) [1] Inf Acceptable? Spencer Graves
Ramon Mart?nez Coscoll? wrote:
We are trying to obtain a predictive distribution in a Poisson model using a non-informative prior distribution. We have found a gamma poisson as a result of our analysis and we want to evaluate it in R. Is there any package where this distribution and anothers are defined? In any case, we have tried to programm the gammapoisson probability function but we have found another problem: we must evaluate a gamma function with a quite high value (gamma(284)). How can it be done? Thank you very much!
______________________________________________ R-help at stat.math.ethz.ch mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help