A semiparametric autorregresive spatial prediction model

TL;DR

This paper introduces a flexible semiparametric spatial autoregressive model combining linear covariates with nonparametric spatial terms, improving interpretability and predictive performance.

Contribution

It develops a novel semiparametric model that relaxes covariance restrictions and provides theoretical guarantees and empirical validation.

Findings

Model achieves competitive predictive accuracy

Provides asymptotic properties like consistency and normality

Enhances interpretability over traditional spatial models

Abstract

In this paper we propose a semiparametric spatial autoregressive model that combines a linear covariate component with a nonparametrically estimated spatial term, allowing flexible dependence modeling without restrictive covariance structure while preserving interpretability. We establish asymptotic properties, including consistency and asymptotic normality, and evaluate performance through simulations and real data. Results show competitive predictive accuracy relative to geostatistical methods and improved interpretability compared to spatial econometric models.