spatemR - Generalized Spatial Autoregresive Models for Mean and Variance
Modeling spatial dependencies in dependent variables,
extending traditional spatial regression approaches. It allows
for the joint modeling of both the mean and the variance of the
dependent variable, incorporating semiparametric effects in
both models. Based on generalized additive models (GAM), the
package enables the inclusion of non-parametric terms while
maintaining the classical theoretical framework of spatial
regression. Additionally, it implements the Generalized Spatial
Autoregression (GSAR) model, which extends classical methods
like logistic Spatial Autoregresive Models (SAR), probit
Spatial Autoregresive Models (SAR), and Poisson Spatial
Autoregresive Models (SAR), offering greater flexibility in
modeling spatial dependencies and significantly improving
computational efficiency and the statistical properties of the
estimators. Related work includes: a) J.D. Toloza-Delgado, Melo
O.O., Cruz N.A. (2024). "Joint spatial modeling of mean and
non-homogeneous variance combining semiparametric SAR and
GAMLSS models for hedonic prices".
<doi:10.1016/j.spasta.2024.100864>. b) Cruz, N. A.,
Toloza-Delgado, J. D., Melo, O. O. (2024). "Generalized spatial
autoregressive model". <doi:10.48550/arXiv.2412.00945>.
