Logo

statsmodels.discrete.discrete_model.Logit.jac

Logit.jac(params)[source]

Logit model Jacobian of the log-likelihood for each observation

Parameters:

params: array-like :

The parameters of the model

Returns:

jac : ndarray, (nobs, k)

The derivative of the loglikelihood evaluated at params for each observation

Notes

\frac{\partial\ln L}{\partial\beta}=\sum_{i=1}^{n}\left(y_{i}-\Lambda_{i}\right)x_{i}

Previous topic

statsmodels.discrete.discrete_model.Logit.initialize

Next topic

statsmodels.discrete.discrete_model.Logit.loglike

This Page