statsmodels.discrete.discrete_model.Logit.hessian¶
method
-
Logit.hessian(params)[source]¶ Logit model Hessian matrix of the log-likelihood
- Parameters
- paramsarray-like
The parameters of the model
- Returns
- hessndarray, (k_vars, k_vars)
The Hessian, second derivative of loglikelihood function, evaluated at params
Notes
\[\frac{\partial^{2}\ln L}{\partial\beta\partial\beta^{\prime}}=-\sum_{i}\Lambda_{i}\left(1-\Lambda_{i}\right)x_{i}x_{i}^{\prime}\]