statsmodels.regression.mixed_linear_model.MixedLM.get_fe_params#

MixedLM.get_fe_params(cov_re, vcomp, tol=1e-10)[source]#

Use GLS to update the fixed effects parameter estimates

Parameters:

cov_rearray_like (2d): The covariance matrix of the random effects.
vcomparray_like (1d): The variance components.
tolfloat, optional: A tolerance parameter to determine when covariances are singular.

Returns:

paramsndarray: The GLS estimates of the fixed effects parameters.
singularbool: True if the covariance is singular

Notes

Directions of the random effects covariance matrix and variance components with an eigenvalue or variance below tol are dropped from the covariance of the observations. The estimates are then the limit of the GLS estimates as these variances go to zero, for example the OLS estimates if all variances are zero.