A misspecification test for the higher order co-moments of the factor model

Wanbo Lu, Dong Yang, Kris Boudt

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

The traditional estimation of higher order co-moments of non-normal random variables by the sample analog of the expectation faces a curse of dimensionality, as the number of parameters increases steeply when the dimension increases. Imposing a factor structure on the process solves this problem; however, it leads to the challenging task of selecting an appropriate factor model. This paper contributes by proposing a test that exploits the following feature: when the factor model is correctly specified, the higher order co-moments of the unexplained return variation are sparse. It recommends a general to specific approach for selecting the factor model by choosing the most parsimonious specification for which the sparsity assumption is satisfied. This approach uses a Wald or Gumbel test statistic for testing the joint statistical significance of the co-moments that are zero when the factor model is correctly specified. The asymptotic distribution of the test is derived. An extensive simulation study confirms the good finite sample properties of the approach. This paper illustrates the practical usefulness of factor selection on daily returns of random subsets of S&P 100 constituents.

Original languageEnglish
Pages (from-to)471-488
JournalStatistics
Volume53
Issue number3
DOIs
Publication statusPublished - 2019

Fingerprint

Misspecification
Factor Models
Higher Order
Moment
Factor Structure
Curse of Dimensionality
Statistical Significance
Sparsity
Asymptotic distribution
Test Statistic
Random variable
Simulation Study
Specification
Analogue
Testing
Subset
Zero
Factors

Keywords

  • 62F03
  • 91B25
  • curse of dimensionality
  • diagnostic test
  • Factor models
  • higher order co-moments
  • sparsity

Cite this

Lu, Wanbo ; Yang, Dong ; Boudt, Kris. / A misspecification test for the higher order co-moments of the factor model. In: Statistics. 2019 ; Vol. 53, No. 3. pp. 471-488.
@article{6d275b6ff544403da19b73450587be38,
title = "A misspecification test for the higher order co-moments of the factor model",
abstract = "The traditional estimation of higher order co-moments of non-normal random variables by the sample analog of the expectation faces a curse of dimensionality, as the number of parameters increases steeply when the dimension increases. Imposing a factor structure on the process solves this problem; however, it leads to the challenging task of selecting an appropriate factor model. This paper contributes by proposing a test that exploits the following feature: when the factor model is correctly specified, the higher order co-moments of the unexplained return variation are sparse. It recommends a general to specific approach for selecting the factor model by choosing the most parsimonious specification for which the sparsity assumption is satisfied. This approach uses a Wald or Gumbel test statistic for testing the joint statistical significance of the co-moments that are zero when the factor model is correctly specified. The asymptotic distribution of the test is derived. An extensive simulation study confirms the good finite sample properties of the approach. This paper illustrates the practical usefulness of factor selection on daily returns of random subsets of S&P 100 constituents.",
keywords = "62F03, 91B25, curse of dimensionality, diagnostic test, Factor models, higher order co-moments, sparsity",
author = "Wanbo Lu and Dong Yang and Kris Boudt",
year = "2019",
doi = "10.1080/02331888.2019.1566907",
language = "English",
volume = "53",
pages = "471--488",
journal = "Statistics",
issn = "0233-1888",
publisher = "Taylor and Francis Ltd.",
number = "3",

}

A misspecification test for the higher order co-moments of the factor model. / Lu, Wanbo; Yang, Dong; Boudt, Kris.

In: Statistics, Vol. 53, No. 3, 2019, p. 471-488.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - A misspecification test for the higher order co-moments of the factor model

AU - Lu, Wanbo

AU - Yang, Dong

AU - Boudt, Kris

PY - 2019

Y1 - 2019

N2 - The traditional estimation of higher order co-moments of non-normal random variables by the sample analog of the expectation faces a curse of dimensionality, as the number of parameters increases steeply when the dimension increases. Imposing a factor structure on the process solves this problem; however, it leads to the challenging task of selecting an appropriate factor model. This paper contributes by proposing a test that exploits the following feature: when the factor model is correctly specified, the higher order co-moments of the unexplained return variation are sparse. It recommends a general to specific approach for selecting the factor model by choosing the most parsimonious specification for which the sparsity assumption is satisfied. This approach uses a Wald or Gumbel test statistic for testing the joint statistical significance of the co-moments that are zero when the factor model is correctly specified. The asymptotic distribution of the test is derived. An extensive simulation study confirms the good finite sample properties of the approach. This paper illustrates the practical usefulness of factor selection on daily returns of random subsets of S&P 100 constituents.

AB - The traditional estimation of higher order co-moments of non-normal random variables by the sample analog of the expectation faces a curse of dimensionality, as the number of parameters increases steeply when the dimension increases. Imposing a factor structure on the process solves this problem; however, it leads to the challenging task of selecting an appropriate factor model. This paper contributes by proposing a test that exploits the following feature: when the factor model is correctly specified, the higher order co-moments of the unexplained return variation are sparse. It recommends a general to specific approach for selecting the factor model by choosing the most parsimonious specification for which the sparsity assumption is satisfied. This approach uses a Wald or Gumbel test statistic for testing the joint statistical significance of the co-moments that are zero when the factor model is correctly specified. The asymptotic distribution of the test is derived. An extensive simulation study confirms the good finite sample properties of the approach. This paper illustrates the practical usefulness of factor selection on daily returns of random subsets of S&P 100 constituents.

KW - 62F03

KW - 91B25

KW - curse of dimensionality

KW - diagnostic test

KW - Factor models

KW - higher order co-moments

KW - sparsity

UR - http://www.scopus.com/inward/record.url?scp=85060208679&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060208679&partnerID=8YFLogxK

U2 - 10.1080/02331888.2019.1566907

DO - 10.1080/02331888.2019.1566907

M3 - Article

VL - 53

SP - 471

EP - 488

JO - Statistics

JF - Statistics

SN - 0233-1888

IS - 3

ER -