Category

Number of Access this item:955（2024-04-26 00:21 Counts）

Identifier to cite or link to this item: https://hdl.handle.net/10086/18594

Link to primary information 

File Format Size Views Description
cover_18594 pdf
gd10-144 pdf 417 KB 815

Item Information

Output File

Title
Model Selection Criteria in Multivariate Models with Multiple Structural Changes
Creator
Creator Name
Kurozumi, Eiji Kurozumi, Eiji
黒住, 英司 黒住, 英司
クロズミ, エイジ クロズミ, エイジ
Affiliation
Affiliation Name
Hitotsubashi University
Creator Name
Tuvaandorj, Purevdorj Tuvaandorj, Purevdorj
Affiliation
Affiliation Name
McGill University
Editor
Contributor Name
社会科学の高度統計・実証分析拠点構築 = Research Unit for Statistical and Empirical Analysis in Social Sciences
Affiliation
Affiliation Name
Hitotsubashi University
Key Word
structural breaks
AIC
Mallows' Cp
BIC
information criteria
JEL
C13
C32
Abstract
This paper considers the issue of selecting the number of regressors and the number of structural breaks in multivariate regression models in the possible presence of multiple structural changes. We develop a modified Akaike's information criterion (AIC), a modified Mallows' Cp criterion and a modified Bayesian information criterion (BIC). The penalty terms in these criteria are shown to be different from the usual terms. We prove that the modified BIC consistently selects the regressors and the number of breaks whereas the modified AIC and the modified Cp criterion tend to overly choose them with positive probability. The finite sample performance of these criteria is investigated through Monte Carlo simulations and it turns out that our modification is successful in comparison to the classical model selection criteria and the sequential testing procedure with the robust method.
Publisher
Institute of Economic Research, Hitotsubashi University
Issued Date
2010-06
Language
English(eng)
Resource Type
technical report
Version Type
VoR
Is Part OfRelation -
Related Title
Global COE Hi-Stat Discussion Paper Series ; No. 144
Number of Pages
45
Sponsorship
グローバルCOEプログラム = Global COE Program
Appears in Collections
Global COE Hi-Stat Discussion Paper Series
URL
https://hdl.handle.net/10086/18594