bayespecon.diagnostics.bayesian_lmtests.bayesian_lm_wx_test

bayespecon.diagnostics.bayesian_lmtests.bayesian_lm_wx_test(model)[source]

Bayesian LM test for WX coefficients (H₀: γ = 0).

Tests whether spatially lagged covariates (WX) should be added to a SAR model, i.e., whether the SAR model should be extended to an SDM specification. Follows the multi-parameter Bayesian LM test framework of Doğan et al. [2021] and the classical LM-WX test of Koley and Bera [2024].

The null model is SAR (includes ρ but not γ). For each posterior draw from the SAR model, residuals are:

\[\mathbf{e} = \mathbf{y} - \rho W \mathbf{y} - X \beta\]

The score vector for the WX coefficients is:

\[\mathbf{g}_\gamma = (WX)^\top \mathbf{e}\]

a \(k_{wx} \times 1\) vector for each draw. The concentration matrix is the information matrix evaluated at the posterior mean:

\[J_{\gamma\gamma} = \frac{1}{\bar{\sigma}^2} (WX)^\top (WX)\]

where \(\bar{\sigma}^2\) is the posterior mean of \(\sigma^2\). The LM statistic for each draw is:

\[\mathrm{LM} = \mathbf{g}_\gamma^\top J_{\gamma\gamma}^{-1} \mathbf{g}_\gamma\]

which is distributed as \(\chi^2_{k_{wx}}\) under H₀.

Parameters:
model : SAR

Fitted SAR model instance with inference_data attribute containing posterior draws for beta, rho, and sigma.

Returns:

Dataclass containing LM samples, summary statistics, and metadata. The df field is set to \(k_{wx}\) (number of WX columns).

Return type:

BayesianLMTestResult