statsmodels.genmod.generalized_linear_model.GLMResults.pseudo_rsquared

GLMResults.pseudo_rsquared(kind='cs')[source]

Pseudo R-squared

Cox-Snell likelihood ratio pseudo R-squared is valid for both discrete and continuous data. McFadden’s pseudo R-squared is only valid for discrete data.

Cox & Snell’s pseudo-R-squared: 1 - exp((llnull - llf)*(2/nobs))

McFadden’s pseudo-R-squared: 1 - (llf / llnull)

Parameters:
kindP”cs”, “mcf”}

Type of pseudo R-square to return

Returns:
float

Pseudo R-squared


Last update: Dec 16, 2024