【Metrics Seminar】Best linear and quadratic moments for spatial econometric models and an application to spatial interdependence patterns of employment growth in US counties
研討會日期 : 2023-01-16
時間 : 14:30
主講人 : Professor Lung-Fei Lee (李龍飛)
地點 : Conference room B110
主持人 : Professor Yu-Chin Hsu
演講者簡介 : Professor Lung-Fei Lee received his Ph.D. in Economics from the University of Rochester in 1977. He is currently an University Chaired Professor at Ohio State University. His research interests are microeconometrics and theoretical econometrics.
演講摘要 : We provide a novel analytic procedure to construct best linear and quadratic moments of generalized method of moments (GMM) estimation for a large class of network and spatial econometric models, which generate a GMM estimator that is asymptotically more efficient than the quasi maximum likelihood estimator when the disturbances are non-normal. We apply this procedure to a high order spatial autoregressive (SAR) model with spatial errors, where the disturbances are heteroskedastic with unknown distribution. Normality tests are proposed for this model. Best moments are also derived for a general high order simultaneous equations SAR model with possible high order multivariate SAR and moving average disturbances, which may have heteroskedastic variances and nests many models in the literature. We apply the high order SAR model and our GMM estimator to local employment data in US counties, which demonstrates spatial interdependence patterns and channels of regional economic growth.