Jump to content

Talk:Generalized gamma distribution

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Lognormal?

[edit]

I do not see how the lognormal could be a special case of the generalized gamma distribution, without any approximation. This should be corrected.--193.50.110.41 (talk) 07:47, 23 June 2010 (UTC)[reply]

I'm sure it's correct, though i'm not sure what the parameters are in the parameterisation shown here. In another parameterisation the lognormal appears as the limit as one of the parameters tends to zero. There are several parameterisations, unfortunately, some simpler for algebra, another better for estimation. I'll try to edit the article when i have time to look it up and sort out how they relate, which probably won't be for a few days... Qwfp (talk) 21:40, 23 June 2010 (UTC)[reply]
According to Crooks [1] (table 1, on page 3), the log-normal distribution can be obtained from the (shifted) generalized gamma distribution in the limit p -> 0, with d = 1/σ2. See also section 3 on page 20. Jheald (talk) 20:57, 27 October 2012 (UTC)[reply]

PDF parameterization

[edit]

I'm not sure the propability density function for this article is correct, nor the graphs that come with it. I made a desmos calculator to show this: https://www.desmos.com/calculator/ua66qlhgbz

The greatest evidence for this is that the article says that when p = 1, the general gamma dist. becomes a regular gamma distribution, but this isn't the case.

What would be the best parameterization for this distribution? Is the propability density function that's here already correct? Thunderblood101 (talk) 16:22, 16 August 2023 (UTC)[reply]