neighbayes.dgp.simulate_sar_probit¶
-
neighbayes.dgp.simulate_sar_probit(W=
None, gdf=None, n_side=None, rho=0.35, beta=None, err_hetero=False, rng=None, seed=None, contiguity='queen', create_gdf=False, geometry_type='polygon', latent=True, *, unit_effects=None)[source]¶ Simulate the SAR probit that
SARProbitfits.DGP¶
y* = (I - rho W)^{-1} (X beta + eps),epsiid standard normal,y = 1(y* > 0): the latent SAR, the unmarked SAR probit. Withlatent=False, the reduced formSARProbit(latent=False)fits:eta = (I - rho W)^{-1} X beta,y ~ Bernoulli(Phi(eta)).- param W:
As for
simulate_sar_logit().- param gdf:
As for
simulate_sar_logit().- param n_side:
As for
simulate_sar_logit().- param contiguity:
As for
simulate_sar_logit().- param rho:
As for
simulate_sar_logit().- param beta:
As for
simulate_sar_logit().- param err_hetero:
As for
simulate_sar_logit().- param rng:
As for
simulate_sar_logit().- param seed:
As for
simulate_sar_logit().- param create_gdf:
As for
simulate_sar_logit().- param geometry_type:
As for
simulate_sar_logit().- param latent:
Falsedraws the reduced form;eta_trueis the latent utilityy*for the latent SAR and the linear predictor for the reduced form.- type latent:
bool, default=True
- returns:
As for
simulate_sar_logit().- rtype:
dict or geopandas.GeoDataFrame