Competitive Selection and Market Data: The Mixed-Index Problem

Roger J. Bowden

The Review of Economic Studies, 1992, vol. 59, issue 3, 625-633

Abstract: If prices have an allocational role, then in addition to their economic signalling, they may act as a statistical signal for the existence of selection bias in cross-sectional studies with some form of price (wages, rental, yield etc.) as dependent variable. Elements of competitive selection appear in the pricing theory of real or financial assets and in the determination of wage rates, where prices correlate the characteristics of the asset or job with those of the individuals, under the equilibrium allocation. If prices achieve such an allocation, then regressions with price as dependent variable are prima facie "mixed-index" in character, with adverse bias or consistency properties, no matter that the right-hand regressors are apparently exogenous. We set up a paradigm for competitive selection and use the concepts so generated to codify the conditions under which mixed index bias will occur. The problem may be regarded as one of selection bias, but does not necessarily derive from classical simultaneity, and is a problem of data-generation structures rather than sample selection.

Date: 1992
References: Add references at CitEc
Citations: View citations in EconPapers (8)

Downloads: (external link)
http://hdl.handle.net/10.2307/2297868 (application/pdf)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:oup:restud:v:59:y:1992:i:3:p:625-633.

Access Statistics for this article

The Review of Economic Studies is currently edited by Thomas Chaney, Xavier d’Haultfoeuille, Andrea Galeotti, Bård Harstad, Nir Jaimovich, Katrine Loken, Elias Papaioannou, Vincent Sterk and Noam Yuchtman

More articles in The Review of Economic Studies from Review of Economic Studies Ltd
Bibliographic data for series maintained by Oxford University Press ().