Asset selection using a factor model and data envelopment analysis– A quantile regression approach

David E. Allen, Abhay Kumar Singh, Robert J. Powell

Research output: Chapter in Book/Report/Conference proceedingChapterResearchpeer-review

Abstract

Given the growing number of stocks and other financial instruments available in the investment market, there is always a need for quick and efficient methods of asset selection for investment purposes. The Fama–French three-factor model has the allure of simplifying asset selection by narrowing the number of parameters required to assess risk, but the usual technique of ordinary least squares regression (OLS) used for the estimation of the coefficients or sensitivity to the three factors suffers from the problem of modeling the conditional mean of the distribution inherent to OLS. In this chapter, we use the technique of data envelopment analysis applied to the Fama–French three-factor model to choose stocks from the Dow Jones Industrial Index. We also apply the more robust technique of quantile regression to estimate the coefficients for the factor model and show that the assets selected using this regression method lead to the construction of superior portfolios with higher returns in equally weighted portfolios when contrasted with the outcomes from OLS.
LanguageEnglish
Title of host publicationRethinking valuation and pricing models
Subtitle of host publicationlessons learned from the crisis and future challenges
EditorsCarsten Wehn, Christian Hoppe , Greg N. Gregoriou
Place of PublicationOxford
PublisherElsevier
Chapter27
Pages443–455
Number of pages13
ISBN (Electronic)9780124158757
DOIs
Publication statusPublished - 2013
Externally publishedYes

Fingerprint

Assets
Data envelopment analysis
Quantile regression
Ordinary least squares
Fama-French three-factor model
Coefficients
Regression method
Financial instruments
Modeling
Factors

Cite this

Allen, D. E., Singh, A. K., & Powell, R. J. (2013). Asset selection using a factor model and data envelopment analysis– A quantile regression approach. In C. Wehn, C. Hoppe , & G. N. Gregoriou (Eds.), Rethinking valuation and pricing models: lessons learned from the crisis and future challenges (pp. 443–455). Oxford: Elsevier. https://doi.org/10.1016/B978-0-12-415875-7.00027-0
Allen, David E. ; Singh, Abhay Kumar ; Powell, Robert J. / Asset selection using a factor model and data envelopment analysis– A quantile regression approach. Rethinking valuation and pricing models: lessons learned from the crisis and future challenges. editor / Carsten Wehn ; Christian Hoppe ; Greg N. Gregoriou. Oxford : Elsevier, 2013. pp. 443–455
@inbook{bcaeffffceaa4cf8942cbe329b913010,
title = "Asset selection using a factor model and data envelopment analysis– A quantile regression approach",
abstract = "Given the growing number of stocks and other financial instruments available in the investment market, there is always a need for quick and efficient methods of asset selection for investment purposes. The Fama–French three-factor model has the allure of simplifying asset selection by narrowing the number of parameters required to assess risk, but the usual technique of ordinary least squares regression (OLS) used for the estimation of the coefficients or sensitivity to the three factors suffers from the problem of modeling the conditional mean of the distribution inherent to OLS. In this chapter, we use the technique of data envelopment analysis applied to the Fama–French three-factor model to choose stocks from the Dow Jones Industrial Index. We also apply the more robust technique of quantile regression to estimate the coefficients for the factor model and show that the assets selected using this regression method lead to the construction of superior portfolios with higher returns in equally weighted portfolios when contrasted with the outcomes from OLS.",
author = "Allen, {David E.} and Singh, {Abhay Kumar} and Powell, {Robert J.}",
year = "2013",
doi = "10.1016/B978-0-12-415875-7.00027-0",
language = "English",
pages = "443–455",
editor = "Carsten Wehn and {Hoppe }, Christian and Gregoriou, {Greg N.}",
booktitle = "Rethinking valuation and pricing models",
publisher = "Elsevier",
address = "Netherlands",

}

Allen, DE, Singh, AK & Powell, RJ 2013, Asset selection using a factor model and data envelopment analysis– A quantile regression approach. in C Wehn, C Hoppe & GN Gregoriou (eds), Rethinking valuation and pricing models: lessons learned from the crisis and future challenges. Elsevier, Oxford, pp. 443–455. https://doi.org/10.1016/B978-0-12-415875-7.00027-0

Asset selection using a factor model and data envelopment analysis– A quantile regression approach. / Allen, David E.; Singh, Abhay Kumar; Powell, Robert J.

Rethinking valuation and pricing models: lessons learned from the crisis and future challenges. ed. / Carsten Wehn; Christian Hoppe ; Greg N. Gregoriou. Oxford : Elsevier, 2013. p. 443–455.

Research output: Chapter in Book/Report/Conference proceedingChapterResearchpeer-review

TY - CHAP

T1 - Asset selection using a factor model and data envelopment analysis– A quantile regression approach

AU - Allen, David E.

AU - Singh, Abhay Kumar

AU - Powell, Robert J.

PY - 2013

Y1 - 2013

N2 - Given the growing number of stocks and other financial instruments available in the investment market, there is always a need for quick and efficient methods of asset selection for investment purposes. The Fama–French three-factor model has the allure of simplifying asset selection by narrowing the number of parameters required to assess risk, but the usual technique of ordinary least squares regression (OLS) used for the estimation of the coefficients or sensitivity to the three factors suffers from the problem of modeling the conditional mean of the distribution inherent to OLS. In this chapter, we use the technique of data envelopment analysis applied to the Fama–French three-factor model to choose stocks from the Dow Jones Industrial Index. We also apply the more robust technique of quantile regression to estimate the coefficients for the factor model and show that the assets selected using this regression method lead to the construction of superior portfolios with higher returns in equally weighted portfolios when contrasted with the outcomes from OLS.

AB - Given the growing number of stocks and other financial instruments available in the investment market, there is always a need for quick and efficient methods of asset selection for investment purposes. The Fama–French three-factor model has the allure of simplifying asset selection by narrowing the number of parameters required to assess risk, but the usual technique of ordinary least squares regression (OLS) used for the estimation of the coefficients or sensitivity to the three factors suffers from the problem of modeling the conditional mean of the distribution inherent to OLS. In this chapter, we use the technique of data envelopment analysis applied to the Fama–French three-factor model to choose stocks from the Dow Jones Industrial Index. We also apply the more robust technique of quantile regression to estimate the coefficients for the factor model and show that the assets selected using this regression method lead to the construction of superior portfolios with higher returns in equally weighted portfolios when contrasted with the outcomes from OLS.

U2 - 10.1016/B978-0-12-415875-7.00027-0

DO - 10.1016/B978-0-12-415875-7.00027-0

M3 - Chapter

SP - 443

EP - 455

BT - Rethinking valuation and pricing models

A2 - Wehn, Carsten

A2 - Hoppe , Christian

A2 - Gregoriou, Greg N.

PB - Elsevier

CY - Oxford

ER -

Allen DE, Singh AK, Powell RJ. Asset selection using a factor model and data envelopment analysis– A quantile regression approach. In Wehn C, Hoppe C, Gregoriou GN, editors, Rethinking valuation and pricing models: lessons learned from the crisis and future challenges. Oxford: Elsevier. 2013. p. 443–455 https://doi.org/10.1016/B978-0-12-415875-7.00027-0