neighbayes.diagnostics.lmtests.bayesian_glm_lm_error_test

neighbayes.diagnostics.lmtests.bayesian_glm_lm_error_test(model)[source]

LM test for a latent error process.

\(q^2 / V_q\) with \(q = e^\top W h\): at the GLM null the score test of the latent SEM, \(\eta = X\beta + S(\lambda)\varepsilon\); at a reduced-form SAR null, of a latent error process added to the fitted model (see the module).

Returns:

Per-draw LM samples, summary statistics, df = 1.

Return type:

BayesianLMTestResult