Is there a missing factor? A canonical correlation approach to factor models

Seung Ahn, Stephan Dieckmann, M. Fabricio Perez

Research output: Contribution to journalArticle

Abstract

A common question in asset pricing research is if a finite set of observable variables can completely capture the systematic or common variations in a large number of response variables. This paper provides a new approach to answer this question. A novelty is that common factors are extracted using canonical relations between response variables and observable factors. We show how these factors in combination with tests for the number of factors can be used to evaluate if a given set of macroeconomic and financial variables is sufficient to capture all the systematic variation in the response variables. We illustrate the usefulness of our methods by analyzing the systematic determinants of credit spreads of U.S. corporate bonds.

LanguageEnglish (US)
JournalReview of Financial Economics
DOIs
StateAccepted/In press - Jan 1 2018
Externally publishedYes

Fingerprint

Factors
Canonical correlation
Common factors
Macroeconomic variables
Usefulness
Corporate bonds
Novelty
Pricing research
Credit spreads
Financial variables
Asset pricing

Keywords

  • Canonical correlations
  • Common factors
  • Credit risk
  • Factor analysis

ASJC Scopus subject areas

  • Finance
  • Economics and Econometrics

Cite this

Is there a missing factor? A canonical correlation approach to factor models. / Ahn, Seung; Dieckmann, Stephan; Perez, M. Fabricio.

In: Review of Financial Economics, 01.01.2018.

Research output: Contribution to journalArticle

@article{2b937c8460db4387b4ee32abf7ef9aef,
title = "Is there a missing factor? A canonical correlation approach to factor models",
abstract = "A common question in asset pricing research is if a finite set of observable variables can completely capture the systematic or common variations in a large number of response variables. This paper provides a new approach to answer this question. A novelty is that common factors are extracted using canonical relations between response variables and observable factors. We show how these factors in combination with tests for the number of factors can be used to evaluate if a given set of macroeconomic and financial variables is sufficient to capture all the systematic variation in the response variables. We illustrate the usefulness of our methods by analyzing the systematic determinants of credit spreads of U.S. corporate bonds.",
keywords = "Canonical correlations, Common factors, Credit risk, Factor analysis",
author = "Seung Ahn and Stephan Dieckmann and Perez, {M. Fabricio}",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.rfe.2017.11.002",
language = "English (US)",
journal = "Review of Financial Economics",
issn = "1058-3300",
publisher = "Elsevier Inc.",

}

TY - JOUR

T1 - Is there a missing factor? A canonical correlation approach to factor models

AU - Ahn, Seung

AU - Dieckmann, Stephan

AU - Perez, M. Fabricio

PY - 2018/1/1

Y1 - 2018/1/1

N2 - A common question in asset pricing research is if a finite set of observable variables can completely capture the systematic or common variations in a large number of response variables. This paper provides a new approach to answer this question. A novelty is that common factors are extracted using canonical relations between response variables and observable factors. We show how these factors in combination with tests for the number of factors can be used to evaluate if a given set of macroeconomic and financial variables is sufficient to capture all the systematic variation in the response variables. We illustrate the usefulness of our methods by analyzing the systematic determinants of credit spreads of U.S. corporate bonds.

AB - A common question in asset pricing research is if a finite set of observable variables can completely capture the systematic or common variations in a large number of response variables. This paper provides a new approach to answer this question. A novelty is that common factors are extracted using canonical relations between response variables and observable factors. We show how these factors in combination with tests for the number of factors can be used to evaluate if a given set of macroeconomic and financial variables is sufficient to capture all the systematic variation in the response variables. We illustrate the usefulness of our methods by analyzing the systematic determinants of credit spreads of U.S. corporate bonds.

KW - Canonical correlations

KW - Common factors

KW - Credit risk

KW - Factor analysis

UR - http://www.scopus.com/inward/record.url?scp=85045902583&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85045902583&partnerID=8YFLogxK

U2 - 10.1016/j.rfe.2017.11.002

DO - 10.1016/j.rfe.2017.11.002

M3 - Article

JO - Review of Financial Economics

T2 - Review of Financial Economics

JF - Review of Financial Economics

SN - 1058-3300

ER -