On determination of the number of factors in an approximate factor model

Jinshan Liu; Jiazhu Pan; Qiang Xia; Li Xiao

doi:10.1515/snde-2020-0055

On determination of the number of factors in an approximate factor model

Jinshan Liu, Jiazhu Pan, Qiang Xia, Li Xiao

Mathematics And Statistics

Research output: Contribution to journal › Article › peer-review

Abstract

This paper proposes a ridge-type method for determining the number of factors in an approximate factor model. The new estimator of factor number is obtained by maximizing both the ratio of two adjacent eigenvalues and the cumulative contribution rate of the factors which represents the explanatory power of the common factors for response variables. Our estimator is proved to be as asymptotically consistent as those in (Ahn, S., and A. Horenstein. 2013. "Eigenvalue Ratio Test for the Number of Factors."Econometrica 81: 1203-27). But Monte Carlo simulation experiments show our method has better correct selection rates in finite sample cases. A real data example is given for illustration.

Original language	English
Number of pages	19
Journal	Studies in Nonlinear Dynamics and Econometrics
Early online date	3 Oct 2022
DOIs	https://doi.org/10.1515/snde-2020-0055
Publication status	E-pub ahead of print - 3 Oct 2022

Keywords

approximate factor model
ridge-type method
cumulative contribution rate
eigenvalue ratio
number of factors

Access to Document

10.1515/snde-2020-0055

Embargoed Document

Liu-etal-SNDE-2022-On-determination-of-the-number-of-factors
Accepted author manuscript, 643 KB
Licence: Strath_1
Embargo ends: 3/10/23

Cite this

@article{3ee3411dd28740d1a29b270f745844b7,

title = "On determination of the number of factors in an approximate factor model",

abstract = "This paper proposes a ridge-type method for determining the number of factors in an approximate factor model. The new estimator of factor number is obtained by maximizing both the ratio of two adjacent eigenvalues and the cumulative contribution rate of the factors which represents the explanatory power of the common factors for response variables. Our estimator is proved to be as asymptotically consistent as those in (Ahn, S., and A. Horenstein. 2013. {"}Eigenvalue Ratio Test for the Number of Factors.{"}Econometrica 81: 1203-27). But Monte Carlo simulation experiments show our method has better correct selection rates in finite sample cases. A real data example is given for illustration.",

keywords = "approximate factor model, ridge-type method, cumulative contribution rate, eigenvalue ratio, number of factors",

author = "Jinshan Liu and Jiazhu Pan and Qiang Xia and Li Xiao",

year = "2022",

month = oct,

day = "3",

doi = "10.1515/snde-2020-0055",

language = "English",

journal = "Studies in Nonlinear Dynamics and Econometrics",

issn = "1558-3708",

}

TY - JOUR

T1 - On determination of the number of factors in an approximate factor model

AU - Liu, Jinshan

AU - Pan, Jiazhu

AU - Xia, Qiang

AU - Xiao, Li

PY - 2022/10/3

Y1 - 2022/10/3

N2 - This paper proposes a ridge-type method for determining the number of factors in an approximate factor model. The new estimator of factor number is obtained by maximizing both the ratio of two adjacent eigenvalues and the cumulative contribution rate of the factors which represents the explanatory power of the common factors for response variables. Our estimator is proved to be as asymptotically consistent as those in (Ahn, S., and A. Horenstein. 2013. "Eigenvalue Ratio Test for the Number of Factors."Econometrica 81: 1203-27). But Monte Carlo simulation experiments show our method has better correct selection rates in finite sample cases. A real data example is given for illustration.

AB - This paper proposes a ridge-type method for determining the number of factors in an approximate factor model. The new estimator of factor number is obtained by maximizing both the ratio of two adjacent eigenvalues and the cumulative contribution rate of the factors which represents the explanatory power of the common factors for response variables. Our estimator is proved to be as asymptotically consistent as those in (Ahn, S., and A. Horenstein. 2013. "Eigenvalue Ratio Test for the Number of Factors."Econometrica 81: 1203-27). But Monte Carlo simulation experiments show our method has better correct selection rates in finite sample cases. A real data example is given for illustration.

KW - approximate factor model

KW - ridge-type method

KW - cumulative contribution rate

KW - eigenvalue ratio

KW - number of factors

U2 - 10.1515/snde-2020-0055

DO - 10.1515/snde-2020-0055

M3 - Article

SN - 1558-3708

JO - Studies in Nonlinear Dynamics and Econometrics

JF - Studies in Nonlinear Dynamics and Econometrics

ER -

On determination of the number of factors in an approximate factor model

Abstract

Keywords

Access to Document

Embargoed Document

Fingerprint

Cite this