Correcting for Omitted-Variable and Measurement-Error Bias in Autoregressive Model Estimation with Panel Data

Authors:
P. A. V. B. Swamy;I-Lok Chang;Jatinder S. Mehta;George S. Tavlas
Affiliations:
Statistical Methods Division, Office of Employment and Unemployment Statistics, Bureau of Labor Statistics, Washington, DC 20212, U.S.A.;Department of Mathematics and Statistics, The American University, Washington, DC 20016, U.S.A.;Department of Mathematics, Temple University, Philadelphia, PA 19122, U.S.A.;Director-Advisor, Economics Research Department, Bank of Greece, Athens, Greece
Venue:
Computational Economics
Year:
2003

Citing 1
Cited 2

A Computational Approach to Finding Causal Economic Laws

Computational Economics - Special issue on computational studies at Cambridge

Empirical best linear unbiased prediction in misspecified and improved panel data models with an application to gasoline demand

Computational Statistics & Data Analysis
An efficient method of estimating the true value of a population characteristic from its discrepant estimates

Computational Statistics & Data Analysis

Quantified Score

Hi-index	0.00

Visualization

Abstract

The parameter estimates based on an econometric equation arebiased and can also be inconsistent when relevant regressors areomitted from the equation or when included regressors are measuredwith error. This problem gets complicated when the 'true'functional form of the equation is unknown. Here, we demonstratehow auxiliary variables, called concomitants, can be used to removeomitted-variable and measurement-error biases from the coefficientsof an equation with the unknown `true' functional form. The methodis specifically designed for panel data. Numerical algorithms forenacting this procedure are presented and an illustration is givenusing a practical example of forecasting small-area employment fromnonlinear autoregressive models.