VALID CONFIDENCE INTERVALS IN REGRESSION AFTER VARIABLE SELECTION

Paul Kabaila

Econometric Theory, 1998, vol. 14, issue 04, pages 463-482

Abstract: We consider a linear regression model with regression parameters ( 1,..., p) and error variance parameter 2. Our aim is to find a confidence interval with minimum coverage probability 1 for a parameter of interest 1 in the presence of nuisance parameters ( 2,..., p, 2). We consider two confidence intervals, the first of which is the standard confidence interval for 1 with coverage probability 1 . The second confidence interval for 1 is obtained after a variable selection procedure has been applied to p. This interval is chosen to be as short as possible subject to the constraint that it has minimum coverage probability 1 . The confidence intervals are compared using a risk function that is defined as a scaled version of the expected length of the confidence interval. We show that, subject to certain conditions including that (dimension of response vector) p is small, the second confidence interval is preferable to the first when we anticipate (without being certain) that p / is small. This comparison of confidence intervals is shown to be mathematically equivalent to a corresponding comparison of prediction intervals.

Date: 1998
View citations in EconPapers

Downloads: (external link)
http://journals.cambridge.org/abstract_S0266466698144031 link to article abstract page (text/html)

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: http://EconPapers.repec.org/RePEc:cup:etheor:v:14:y:1998:i:04:p:463-482_14

Access Statistics for this article

More articles in Econometric Theory from Cambridge University Press
Address: The Edinburgh Building, Shaftesbury Road, Cambridge CB2 2RU UK
Series data maintained by Mike Eden ().