Add loggamma

[tirthajyoti]

I have a problem where loggamma fits best. I ran your script and my own custom script, they agree on beta parameters but the loggamma seemed much more natural. If it's not too much trouble, please consider adding this. If you are using scipy.stats, then it's the same API as others.

Cool project.

[erdogant]

Thanks! The loggamma is already part of distfit.
For example:

from distfit import distfit
X = np.random.normal(0, 2, 5000)
y = [-8,-6,0,1,2,3,4,5,6]
dist = distfit(distr='loggamma')
dist.fit_transform(X)
dist.plot()

Is this what you mean?

[tirthajyoti]

Yes, it is what I meant. However, is it not included in your regular default/popular run?

When I ran,

dist = distfit(todf=True)
dist.fit_transform(dataset)
dist.plot()

The result did not include loggamma and the lowest RSS was for the beta which produced large, unnatural parameters. In my case, the loggamma produced a far lower RSS score and would have been selected as the best one in the default run. I guess it's not that popular, so your popular choice does not include it.

You can close the issue now.

[erdogant]

Well, "popular" is not a fixed list. I included loggamma in the list.

Update with:

pip install -U distfit