neighbayes.dgp.simulate_sem_flow

neighbayes.dgp.simulate_sem_flow(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', *, T=None, pair_effects=None, time_effects=False, separable=False, pair_effect_sd=None, time_effect_sd=None)[source]

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

Spatial-error analogue of simulate_sar_flow(). 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 simulate_sar_flow().