Spatial Panel Data Models: Unbalance Panel, Threshold Effect and Network Structure

Title:

Spatial Panel Data Models: Unbalance Panel, Threshold Effect and Network Structure

ABSTRACT

This dissertation studies the estimation and inference problems for spatial panel data (SPD) models when (i) panels are unbalanced, (ii) panels contain threshold effects, or (iii) panels contain time-varying network structures. Chapter 1 concerns the unbalanced SPD models with fixed effects (FE). The unbalanced nature of the panel data renders the standard method of estimation inapplicable. We propose an M-estimation method where the estimating functions are obtained by adjusting the concentrated quasi scores to account for the estimation of FEs. The methods are further extended to allow for the presence of unknown spatiotemporal heteroscedasticity. Consistency and asymptotic normality of the proposed estimators are established. Monte Carlo results show excellent finite sample performance of the proposed estimators and the proposed inference methods. Chapter 2 introduces general estimation and inference methods for threshold SPD models under a diminishing-threshold-effects framework with a balanced panel. A valid objective function is first obtained by a simple adjustment on the concentrated quasi loglikelihood with FEs being concentrated out. We then show that the estimation of the threshold parameter has an asymptotically negligible effect on the asymptotic distribution of the other estimators, and thereby leads to valid inference methods for other common parameters after a bias correction. A likelihood ratio test is proposed for statistical inference on the threshold parameter. We also propose a sup-Wald test for the presence of threshold effects. Monte Carlo results show good finite sample properties of the proposed estimation and inference methods. Chapter 3 considers the specification and estimation of a three-dimensional SPD model with time-varying network structures. The model allows for endogenous and exogenous interaction effects, correlation of unobservables, and most importantly group-specific effects that are allowed to interact with the individual and time specific effects. The unbiased estimating functions are obtained by adjusting the concentrated quasi scores (with FEs being concentrated out), leading to adjusted quasi-score estimators that are consistent and asymptotically normal. Monte Carlo results also show excellent finite sample performance of the proposed method.

PRESENTER

MENG Xiaoyu
PhD Candidate
School of Economics
Singapore Management University

DISSERTATION COMMITTEE:

Chair:
Professor YANG Zhenlin
Professor of Economics & Statistics
Singapore Management University

Committee Member:
Professor Yichong ZHANG
Assistant Professor of Economics
Lee Kong Chian Fellow
Singapore Management University

Professor JIN Sainan
Professor of Economics
Tsinghua University

External Member:
Professor HAN Xiaoyi
Associate Professor of Economics
Xiamen University

RESEARCH FIELDS

Spatial Econometrics, Panel Data Models, Applied Econometrics

DATE:

22 June 2022 (Wednesday)

TIME:

10.00 am

VENUE:

SOE Seminar Room 5.1, Level 5
School of Economics
Singapore Management University
90 Stamford Road
Singapore 178903