Two Papers on Model Testing and Discrimination

Gordon Fisher and Michael McAleer

No 275191, Queen's Institute for Economic Research Discussion Papers from Queen's University - Department of Economics

Abstract: In this note it is demonstrated that Theil's (1961) minimum error variance criterion is asymptotically valid for choosing between non-nested nonlinear regression models, as long as one of the models being considered is 'true'. -- Pereira (1977) raised the issue of consistency of the tests , developed by Cox (1961, 1962) and Atkinson (1970) for separate families of hypotheses and showed that the Cox-test is always consistent whereas Atkinson's procedure is not always consistent. In this paper it is shown that when the null and alternative hypotheses are separate regression models, Atkinson's procedure will always provide a consistent test. It is also shown that, in addition to comparing the actual and expected performances of the alternative when the Coxtest is used, the actual performance of the null becomes relevant for interpreting the outcome of the test based on Atkinson's modification.

Keywords: Financial Economics; Research Methods/ Statistical Methods (search for similar items in EconPapers)
Pages: 22
Date: 1980-12
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://ageconsearch.umn.edu/record/275191/files/QUEENS-IER-PAPER-416.pdf (application/pdf)

Related works:
Working Paper: Two Papers on Model Testing and Discrimination (1980)
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:ags:queddp:275191

DOI: 10.22004/ag.econ.275191

Access Statistics for this paper

More papers in Queen's Institute for Economic Research Discussion Papers from Queen's University - Department of Economics Contact information at EDIRC.
Bibliographic data for series maintained by AgEcon Search ().