No label defined (Q3321263)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 3852237
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 3852237 |
Statements
1983
0 references
deletion of relevant variable
0 references
inclusion of irrelevant variable
0 references
operational generalized least squares method
0 references
seemingly unrelated regression equation model
0 references
ordinary least squares
0 references
two equation model
0 references
specification error
0 references
asymptotic properties
0 references
0.7731472253799438
0 references
0.773098349571228
0 references
0.7696704864501953
0 references
\textit{A. Zellner} [J. Am. Stat. Assoc. 57, 348-368 (1962; Zbl 0113.349)] proposed an operational generalized least squares method (OGLS) for estimating the regression coefficients in a seemingly unrelated regression equation model, which provides asymptotically more efficient estimators than the ordinary least squares (OLS) method. All the results, however, assume that the model contains no specification errors. \textit{P. Rao} [Specification bias in seemingly unrelated regressions. Econometrics and economic theory, Essays in Hon. of J. Tinbergen, 101-113 (1974)] considered a simple two equation model in three variables and investigated the asymptotic properties of OLS and OGLS estimates when a variable common to both the equations is wrongly deleted. The present authors postulated a two equation model with specification error in one equation, which comprises either deletion of a relevant variable or inclusion of an irrelevant one, and they studied the asymptotic properties of OLS and OGLS estimates of the regression coefficients. In some cases, OLS estimates are better. In other cases, OGLS estimates are better.
0 references