Review of Economic Studies Vol. 53 No. 4 1986
Hypothesis Testing in Unidentified Models
Abstract
An identified model is not necessary for statistical inference, but ambiguities can arise. This paper examines some simple examples and proposes a framework that distinguishes between the “refutation” and “confirmation” aspects of testing in an unidentified model. One particular problem is the interpretation given to overidentifying restrictions: a common view is that these are somehow not properly testable.
- DOI
- 10.2307/2297609
- Volume
- 53
- Issue
- 4
- Pages
- 635
- Sources
- bibtex:phds-export.bib openalex crossref