Category

Number of Access this item:927（2024-04-19 08:28 Counts）

Identifier to cite or link to this item: https://hdl.handle.net/10086/18769

Link to primary information 

File Format Size Views Description
cover_18769 pdf
gd10-156 pdf 205 KB 760

Item Information

Output File

Title
Investigating Finite Sample Properties of Estimators for Approximate Factor Models When N Is Small
Creator
Creator Name
Tanaka, Shinya Tanaka, Shinya
Affiliation
Affiliation Name
Hitotsubashi University
Creator Name
Kurozumi, Eiji Kurozumi, Eiji
Affiliation
Affiliation Name
Hitotsubashi 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
Approximate factor model
Principal components,
Quasi-maximum likelihood
JEL
C32
C33
C63
Abstract
This paper examines the finite sample properties of estimators for approximate factor models when N is small via simulation study. Although the “rule-of-thumb” for factor models does not support using approximate factor models when N is small, we find that the principal component analysis estimator and quasi-maximum likelihood estimator proposed by Doz et al. (2008) perform very well even in this case. Our findings provide an opportunity for applying approximate factor models to low-dimensional data, which was thought to have been inappropriate for a long time.
Publisher
Institute of Economic Research, Hitotsubashi University
Issued Date
2010-12
Language
English(eng)
Resource Type
technical report
Version Type
VoR
Is Part OfRelation -
Related Title
Global COE Hi-Stat Discussion Paper Series ; No. 156
Number of Pages
8
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/10086/18769