2014
DOI: 10.1016/j.jeconom.2013.05.007
|View full text |Cite
|
Sign up to set email alerts
|

Model equivalence tests in a parametric framework

Abstract: In empirical research, one commonly aims to obtain evidence in favor of restrictions on parameters, appearing as an economic hypothesis, a consequence of economic theory, or an econometric modeling assumption. I propose a new theoretical framework based on the Kullback-Leibler information to assess the approximate validity of multivariate restrictions in parametric models. I construct tests that are locally asymptotically maximin and locally asymptotically uniformly most powerful invariant. The tests are appli… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
5
0

Year Published

2016
2016
2021
2021

Publication Types

Select...
3
1

Relationship

0
4

Authors

Journals

citations
Cited by 4 publications
(5 citation statements)
references
References 50 publications
0
5
0
Order By: Relevance
“…Let g(θ 0 ) = 0 denote multivariate restrictions on θ 0 with r number of restrictions. As a measure of closeness to the true distribution, Lavergne (2014) adopts the Kullback-Leibler information criterion, which is defined as…”
Section: Model Equivalence Testmentioning
confidence: 99%
See 3 more Smart Citations
“…Let g(θ 0 ) = 0 denote multivariate restrictions on θ 0 with r number of restrictions. As a measure of closeness to the true distribution, Lavergne (2014) adopts the Kullback-Leibler information criterion, which is defined as…”
Section: Model Equivalence Testmentioning
confidence: 99%
“…Rejection of H 0 implies that the restriction g(θ 0 ) = 0 is close to be valid. According to Lavergne (2014), the above model equivalence test can be conducted using the log-likelihood ratio (LR) test, which can be written as…”
Section: Model Equivalence Testmentioning
confidence: 99%
See 2 more Smart Citations
“…It refers to what Berger and Delampady (1987) call the principle of precise hypothesis about a parameter. In the more familiar current context of GOF testing it may also be found in Dette and Munk (1998) where this principle is applied to an L 2 -type distance procedure with an asymptotic normal law; see also Borovkov (1998), §49 &§55, andLavergne (2014). I am not aware of work related to this approach based on the ECF.…”
Section: Bootstrap Critical Valuesmentioning
confidence: 99%