The Use of Error Components Models in Combining Cross Section with Time Series Data
Repository Usage Stats
A mixed model of regression with error components is proposed as one of possible interest for combining cross section and time series data. For known variances, it is shown that Aitken estimators and covariance estimators are in one sense asymptotically equivalent, even though the Aitken estimators are more efficient in small samples. Turning to unknown variance components, Zellner-type iterative estimators are compared with covariance estimators. Here, few small sample properties are obtained. However, it is shown that covariance and Zellner-type estimators have equivalent asymptotic distributions and equivalent limits of sequences of first and second order moments for weakly nonstochastic regressors. For the model analyzed, the theoretical results obtained, as well as ease of computation, tend to support traditional covariance estimators of the regression parameters. An additional interesting result presented in an appendix is that ordinary least squares estimates of the β's (ignoring the error components) have unbounded asymptotic variances. On efficiency grounds, this argues rather strongly for some care in combining data from alternative sources in regression analysis.
More InfoShow full item record
James B. Duke Distinguished Professor Emeritus of Economics
Professor Wallace’s most recent endeavor was the completion of a textbook covering general knowledge within his field. The book was Econometrics: An Introduction, written in collaboration with his former student, Lew Silver. As a researcher, his investigations explored such variables as human capital accumulation, linear restrictions in regression, time series data, multicollinearity and low-order moments in stable lag distribution, fertility and replacement, full time schooling, the mean squa
This author no longer has a Scholars@Duke profile, so the information shown here reflects their Duke status at the time this item was deposited.