Re: [R] logLik for poisson models

On Mon, 2 Feb 2009, antonio.gasparr...@lshtm.ac.uk wrote: Dear all, I have a very basic question: how does the logLik function work for poisson models? You mean its inner workings or just what is the computation? The computation is just the vanilla loglikelihood: sum( dpois( y, mean

Re: [R] logLik for poisson models

Antonio, Given the model you used, the log link means you are modeling log(expected value of y)=intercept If you had an independent variable e.g. x model <- glm(y ~ x, family=poisson()) you would be modeling log(expected value of y)=intercept+x John John David Sorkin M.D., Ph.D. Chief, Biosta

[R] logLik for poisson models

Dear all, I have a very basic question: how does the logLik function work for poisson models? Example: I simulate 20 observations from a Poisson distribution with mean 800. y <- rpois(20,800) model <- glm(y ~ 1, family=poisson()) logLik(model) I would like to know what's the exact formula th