statsmodels.discrete.discrete_model.Poisson.score_factor¶
method
-
Poisson.score_factor(params)[source]¶ Poisson model score_factor for each observation
- Parameters
- paramsarray-like
The parameters of the model
- Returns
- scorearray-like
The score factor (nobs, ) of the model evaluated at params
Notes
\[\frac{\partial\ln L_{i}}{\partial\beta}=\left(y_{i}-\lambda_{i}\right)\]for observations \(i=1,...,n\)
where the loglinear model is assumed
\[\ln\lambda_{i}=x_{i}\beta\]