JuliaStats / GLM.jl

Generalized linear models in Julia

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Feature Request: Post Estimation Score & InformationMatrix

azev77 opened this issue · comments

It is often useful to have the score & information matrix of an estimated model (for example to do a score test).
Julia StatsBase allows StatsBase.informationmatrix(m) & StatsBase.score(m).
It would be awesome if GLM implemented these.

Sounds like a great first PR!

Is there a way to access the loglikelihood as a function of the parameters as each datapoint? Something like loglik_obs that maps from parameters and obs to a number.

It would be also useful to have access to some other functions available in statsmodels in python: statsmodels.glmmodel.score_factor and statsmodels.glmmodel.hessian_factor