What are the distribution functions of the gamma distribution shape and scale parameters?

If I compute the shape and scale parameters of the gamma distribution, I would like to known their accuracies, which I could estimate if I knew the probability distributions of each parameter.

[José-Luis]

What is it you are trying to do? Your question is not clear.

[Richard McCuen]

I want to transform the probability distributions (pdrs) of the two parameters to determine the pdfs of transformed variables.

Answer 1

[Torsten]

Use gamfit. It gives you confidence intervals for the estimated parameters.

Best wishes

Torsten.

[Richard McCuen]

I need more than the confidence interval. I need to know the distribution functions used to compute the CIs.

[José-Luis]

According to the documentation, it's MLE confidence intervals. How they are computed is a longish story but the short and dirty is that they come from a normal distribution (sort of).

[John D'Errico]

Exactly. Nothing is truly normal, always something else (and messy) that is very hard for which to write down an exact distribution. But all sorts of things are asymptotically normal. Those confidence intervals were computed using normality as the presumption.