Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
7358039 | Journal of Econometrics | 2018 | 41 Pages |
Abstract
This paper develops an innovative way of estimating a functional-coefficient spatial autoregressive panel data model with unobserved individual effects which can accommodate (multiple) time-invariant regressors with a large number of cross-sectional units and a finite time periods. Our proposed methodology removes unobserved fixed effects from the model by transforming the latter into a semiparametric additive model, however avoids using backfitting technique. We derive the limiting results for the proposed estimators and construct a consistent nonparametric test to test for spatial endogeneity. A small Monte Carlo study shows that our proposed estimators and test statistic exhibit good finite-sample performance.
Keywords
Related Topics
Physical Sciences and Engineering
Mathematics
Statistics and Probability
Authors
Yiguo Sun, Emir Malikov,