Advanced

On the determination of the number of factors using information criteria with data-driven penalty

Westerlund, Joakim LU and Mishra, Sagarika (2017) In Statistical Papers 58(1). p.161-184
Abstract

As is well known, when using an information criterion to select the number of common factors in factor models the appropriate penalty is generally indetermine in the sense that it can be scaled by an arbitrary constant, c say, without affecting consistency. In an influential paper, Hallin and Liška (J Am Stat Assoc102:603–617, 2007) proposes a data-driven procedure for selecting the appropriate value of c. However, by removing one source of indeterminacy, the new procedure simultaneously creates several new ones, which make for rather complicated implementation, a problem that has been largely overlooked in the literature. By providing an extensive analysis using both simulated and real data, the current paper fills this gap.

Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Common factor model, Data-driven penalty, Information criterion, Panel data
in
Statistical Papers
volume
58
issue
1
pages
24 pages
publisher
Springer
external identifiers
  • scopus:84930348545
  • wos:000394997000009
ISSN
0932-5026
DOI
10.1007/s00362-015-0692-0
language
English
LU publication?
yes
id
bc0fb0f2-d854-40f5-9b9e-54f102046980
date added to LUP
2017-03-13 09:28:28
date last changed
2018-01-07 11:55:02
@article{bc0fb0f2-d854-40f5-9b9e-54f102046980,
  abstract     = {<p>As is well known, when using an information criterion to select the number of common factors in factor models the appropriate penalty is generally indetermine in the sense that it can be scaled by an arbitrary constant, c say, without affecting consistency. In an influential paper, Hallin and Liška (J Am Stat Assoc102:603–617, 2007) proposes a data-driven procedure for selecting the appropriate value of c. However, by removing one source of indeterminacy, the new procedure simultaneously creates several new ones, which make for rather complicated implementation, a problem that has been largely overlooked in the literature. By providing an extensive analysis using both simulated and real data, the current paper fills this gap.</p>},
  author       = {Westerlund, Joakim and Mishra, Sagarika},
  issn         = {0932-5026},
  keyword      = {Common factor model,Data-driven penalty,Information criterion,Panel data},
  language     = {eng},
  month        = {03},
  number       = {1},
  pages        = {161--184},
  publisher    = {Springer},
  series       = {Statistical Papers},
  title        = {On the determination of the number of factors using information criteria with data-driven penalty},
  url          = {http://dx.doi.org/10.1007/s00362-015-0692-0},
  volume       = {58},
  year         = {2017},
}