Download presentation
Presentation is loading. Please wait.
Published byAlexandrina Rice Modified over 8 years ago
1
Spatial Econometric Analysis 3 Kuan-Pin Lin Portland State University
2
Model Estimation Spatial Lag Model SPLAG(1) OLS is biased and inconsistent.
3
Spatial Lag Model IV or 2SLS Estimation Instrumental Variables Two-Stage Least Squares
4
Spatial Lag Model IV/2SLS with SHAC
5
Spatial Lag Model GMM Estimation Strong Exogeneity of Instrumental Variables Generalized Method of Moments (GMM)
6
Spatial Lag Model GMM Estimation Efficient GMM Estimator
7
Spatial Lag Model Maximum Likelihood Estimation Normal Density Function
8
Spatial Lag Model Maximum Likelihood Estimation Jacobian Matrix
9
Spatial Lag Model Maximum Likelihood Estimation Log-Likelihood Function
10
Spatial Lag Model Maximum Likelihood Estimation Quasi Maximum Likelihood (QML) Estimator
11
Crime Equation Anselin (1988) Spatial Lag Model (Crime Rate) = + (Family Income) + (Housing Value) + W (Crime Rate) + OLS vs. IV Estimator OLS Parameter OLS s.e. IV Parameter IV s.e 0.557330.150290.454560.17440 -0.865840.35541-1.00070.36786 -0.263850.09136-0.26550.08802 38.1819.215343.79410.496 R2R2 0.65720.6536
12
Crime Equation Anselin (1988) Spatial Lag Model (Crime Rate) = + (Family Income) + (Housing Value) + W (Crime Rate) + IV with SHAC Estimator IV Parameter IV s.e IV s.e./hc IV s.e/hac 0.454560.174400.142590.17428 -1.00070.367860.456300.48617 -0.26550.088020.173740.17306 43.79410.4967.75798.2884 R2R2 0.6536
13
Crime Equation Anselin (1988) Spatial Lag Model (Crime Rate) = + (Family Income) + (Housing Value) + W (Crime Rate) + GMM Estimator GMM-hc Parameter GMM-hc s.e GMM-hac Parameter GMM-hac s.e 0.420170.129620.420230.13177 -1.18910.42062-0.976830.40565 -0.216250.15437-0.274280.14808 46.2567.066844.9226.9797 R2R2 0.64740.6504
14
Crime Equation Anselin (1988) Spatial Lag Model (Crime Rate) = + (Family Income) + (Housing Value) + W (Crime Rate) + QML vs. GMM Estimator QML Parameter QML s.e GMM-hac Parameter GMM-hac s.e 0.431010.129620.420230.13177 -1.03160.42108-0.976830.40565 -0.265930.17309-0.274280.14808 45.0806.405144.9226.9797 L-182.39-181.1
15
References Kelejian , H. , and I. R. Prucha , 1999. A Generalized Moments Estimator for the Autoregressive Parameter in a Spatial Model. International Economic Review 40 , 509-533. Kelejian, H., and I. R. Prucha, 2009. Specification and Estimation of Spatial Autoregressive Models with Autoregressive and Heteroskedastic Disturbances. Journal of Econometrics, forthcoming. Lee , L. F. , 2004. Asymptotic Distributions of Maximum Likelihood Estimators for Spatial Autoregressive Models. Econometrica, 72, 1899-1925.
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.