Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/253585 
Year of Publication: 
2020
Citation: 
[Journal:] Quantitative Economics [ISSN:] 1759-7331 [Volume:] 11 [Issue:] 3 [Publisher:] The Econometric Society [Place:] New Haven, CT [Year:] 2020 [Pages:] 983-1017
Publisher: 
The Econometric Society, New Haven, CT
Abstract: 
This paper proposes a new model selection test for the statistical comparison of semi/non-parametric models based on a general quasi-likelihood ratio criterion. An important feature of the new test is its uniformly exact asymptotic size in the overlapping nonnested case, as well as in the easier nested and strictly nonnested cases. The uniform size control is achieved without using pretesting, sample-splitting, or simulated critical values. We also show that the test has nontrivial power against all vn-local alternatives and against some local alternatives that converge to the null faster than vn. Finally, we provide a framework for conducting uniformly valid post model selection inference for model parameters. The finite sample performance of the nondegenerate test and that of the post model selection inference procedure are illustrated in a mean-regression example by Monte Carlo.
Subjects: 
Asymptotic size
model selection/comparison test
post model selection inference
semi/nonparametric models
JEL: 
C14
C31
C32
Persistent Identifier of the first edition: 
Creative Commons License: 
cc-by-nc Logo
Document Type: 
Article

Files in This Item:
File
Size
553.02 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.