0
$\begingroup$

Can anybody show me, how to find a CI to Dispersion parameter of negbin distribution in R?

Iknow how to calculate the point estimate but I do not know how to find the confidence interval

so, what is the IC for this example: enter link description here

$\endgroup$
1
$\begingroup$

The theta parameter is not part of the main model parameters in glm.nb (as extracted by coef()) but it is treated as a nuisance parameter. However, the fitted model object contains the parameter estimate ($theta) and its standard error ($SE.theta). So you can easily employ a Wald confidence interval based on the asymptotic normal distribution (essentially +/- two standard errors).

library("MASS")
a <- glm.nb(Days ~ Eth + Sex, data = quine)
a$theta + qnorm(c(0.025, 0.975)) * a$SE.theta
## [1] 0.8872956 1.4555238
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.