A nonparametric spatial regression model using partitioning estimators

Olmo, Jose ; Sanso-Navarro, Marcos (Universidad de Zaragoza)

Resumen: Conventional spatial regression models are extended by modelling the spatial effects of the exogenous regressor model (SLX) as a functional coefficient. This coefficient is estimated by partitioning the domain of the spatial variable into a set of disjoint intervals and approximating the function using local Taylor expansions. The asymptotic properties of the proposed partitioning estimator are derived, and pointwise and uniform tests for the presence of spatial effects are developed. An empirical application of this work is used to study environmental Engel curves and provides strong evidence of neighbouring effects in the relationship between households’ income and the amount of pollution embodied in the goods and services they consume.

Idioma: Inglés

DOI: 10.1016/j.ecosta.2023.02.003

Año: 2026

Publicado en: Econometrics and Statistics 37 (2026), 126-153
ISSN: 2452-3062

Factor impacto CITESCORE: 3.1 - Economics and Econometrics (Q2) - Statistics, Probability and Uncertainty (Q2) - Statistics and Probability (Q2)

Factor impacto SCIMAGO: 0.69 - Economics and Econometrics (Q2) - Statistics, Probability and Uncertainty (Q2) - Statistics and Probability (Q2)

Financiación: info:eu-repo/grantAgreement/ES/DGA/LMP21-21
Financiación: info:eu-repo/grantAgreement/ES/DGA/S39-20R
Financiación: info:eu-repo/grantAgreement/ES/MICINN/PID2019-104326GB-I00

Tipo y forma: Article (PostPrint)

Área (Departamento): Área Fund. Análisis Económico (Dpto. Análisis Económico)