Open fkiraly opened 1 month ago
Regarding point 1, architecture, I would propose a delegator design, using the _DelegatedDistribution
:
Tweedie
can be a delegator, delegating to the subfamily determined by p
Regarding point 3, section 3 of Withers, Nadarajah - "On the compound Poisson-gamma distribution" has numerically exact approximation formulae for cdf and ppf, for the CPG.
@ShreeshaM07, could you kindly reply to this so I can assign you?
Sure. So I have been working on the implementation of the Tweedie Distribution. So far completed the pdf and cdf implementation of the sub distribution which is the compound poisson gamma distribution in #428.
For https://github.com/sktime/skpro/issues/423 and other GLM interfaces, we would need a Tweedie distribution.
The Tweedie distribution has a number of difficulties:
p
, a real number, which leads to special cases that include well-known and obscure distribution families. This also leads to design questions on how these families and Tweedie should relatescipy
interface, hence it does not exist inscipy
for allp
. In fact,scipy
has no Tweedie distribution, see discussion here: https://github.com/scipy/scipy/issues/11291#issuecomment-1868256070cdf
orppf
There is some discussion about Tweedie in the issue related to the
sklearn
Tweedie regressor: https://github.com/sktime/skpro/issues/423#issuecomment-2231637334