Econometric Analysis of Panel Data William Greene Department of Economics Stern School of Business
The Hausman and Taylor RE Model
A Useful Result: LSDV is an IV Estimator
Hausman and Taylor Use that Result
H&T’s FGLS Estimator
H&T’s FGLS Estimator (cont.)
H&T’s 4 STEP IV Estimator
The Hausman and Taylor Application: NLSY Wage Equation
Standard Textbook Application Based on Cornwell and Rupert
Is marital status really endogenous?
Dynamic (Linear) Panel Data (DPD) Models Application Bias in Conventional Estimation Development of Consistent Estimators Efficient GMM Estimators
THE Dynamic Linear Model
A General DPD model
OLS and GLS are inconsistent
LSDV is Also Inconsistent [(Steven) Nickell Bias]
IV Estimation of the DPD Model Anderson Hsiao IV Estimator
Arellano and Bond Estimator - 1
Arellano and Bond Estimator - 2
Ahn and Schmidt
Arellano/Bond/Bover’s Formulation Start with H&T
Arellano/Bond/Bover’s Formulation Dynamic Model
Arellano/Bond/Bover’s Formulation
Arellano/Bond/Bover Estimator
GMM Criterion
Application: Maquiladora
Maquiladora
Postscript There is no theoretical guidance on the instrument set There is no theoretical guidance on the form of the covariance matrix There is no theoretical guidance on the number of lags at any level of the model There is no theoretical guidance on the form of the exogeneity – and it is not testable. Results vary wildly with small variations in the assumptions.
Appendix
The Panel Data Case
GMM Estimation for One Equation
GMM for a System of Equations
SUR Model with Endogenous RHS Variables
GMM for the System - Notation
Instruments
Moment Equations
Estimation-1
Estimation-2
Estimation-3
Estimation-4
Estimation-5
Arellano and Bond Estimator - 3
Arellano and Bover Instrumental Variables Approach
Simple IV Estimation
Arellano/Bond/Bover’s Formulation These blocks may contain all previous exogenous variables, or all exogenous variables for all periods. This may contain the all periods of data on x1 rather than just the group mean. (Amemiya and MaCurdy).
Arellano/Bond/Bover’s Formulation
Arellano/Bond First Difference Formulation
Arellano/Bond - GLS
Arellano/Bond GLS Estimator
GMM Estimator
http://people. stern. nyu http://people.stern.nyu.edu/wgreene/CumulantInstruments-Racicot-AE(2014)_46(10).pdf
The NYU No Action Letter
Arellano/Bond/Bover’s Formulation For unbalanced panels the number of columns for Zi varies. Given the form of Zi, the number of columns depends on Ti. We need all Zi to have the same number of columns. For matrices with less columns than the largest one, extra columns of zeros are added.