bayespecon.dgp.generate_sem_flow_data¶

bayespecon.dgp.generate_sem_flow_data(n=None, G=None, lam_d=0.3, lam_o=0.2, lam_w=0.1, beta_d=None, beta_o=None, sigma=1.0, X=None, col_names=None, dist=None, gamma_dist=-0.5, alpha=0.0, seed=None, gdf=None, err_hetero=False, knn_k=4, distribution='lognormal')[source]¶

Simulate flow data from a spatial-error (SEM) flow model.

Spatial-error analogue of generate_flow_data(). The latent additive linear predictor is

\[\eta = X\beta + B^{-1} \varepsilon, \quad B = I_N - \lambda_d W_d - \lambda_o W_o - \lambda_w W_w, \quad \varepsilon \sim \mathcal{N}(0, \sigma^2 I_N)\]

with observed flows \(y = \exp(\eta)\) (default "lognormal") or \(y = \eta\) ("normal"). Use to generate training data for SEMFlow.

Parameters and return dict mirror generate_flow_data().