statsmodels.genmod.families.family.Tweedie.loglike_obs

Tweedie.loglike_obs(endog, mu, var_weights=1.0, scale=1.0)[source]

The log-likelihood function for each observation in terms of the fitted mean response for the Tweedie distribution.

Parameters:

  • endog (ndarray) – Usually the endogenous response variable.

  • mu (ndarray) – Usually but not always the fitted mean response variable.

  • var_weights (array_like) – 1d array of variance (analytic) weights. The default is 1.

  • scale (float) – The scale parameter. The default is 1.

Returns:

ll_i – The value of the loglikelihood evaluated at (endog, mu, var_weights, scale) as defined below.

Return type:

float

Notes

If eql is True, the Extended Quasi-Likelihood is used. At present, this method returns NaN if eql is False. When the actual likelihood is implemented, it will be accessible by setting eql to False.

References

R Kaas (2005). Compound Poisson Distributions and GLM’s – Tweedie’s Distribution. https://core.ac.uk/download/pdf/6347266.pdf#page=11

JA Nelder, D Pregibon (1987). An extended quasi-likelihood function. Biometrika 74:2, pp 221-232. https://www.jstor.org/stable/2336136