Portfolio performance of linear SDF models: an out-of-sample assessment

Martín Carlos Lozano Banda, Edwin Hansen, Massimo Guidolin

Research output: Contribution to journalArticle

Abstract

We evaluate linear stochastic discount factor models using an ex-post portfolio metric: the realized out-of-sample Sharpe ratio of mean–variance portfolios backed by alternative linear factor models. Using a sample of monthly US portfolio returns spanning the period 1968–2016, we find evidence that multifactor linear models have better empirical properties than the CAPM, not only when the cross-section of expected returns is evaluated in-sample, but also when they are used to inform one-month ahead portfolio selection. When we compare portfolios associated to multifactor models with mean–variance decisions implied by the single-factor CAPM, we document statistically significant differences in Sharpe ratios of up to 10 percent. Linear multifactor models that provide the best in-sample fit also yield the highest realized Sharpe ratios.
Original languageEnglish
Pages (from-to)1425-1436
Number of pages12
JournalQuantitative Finance
Volume18
Issue number8
DOIs
Publication statusPublished - 3 Aug 2018

Fingerprint

Portfolio performance
Sharpe ratio
Capital asset pricing model
Multifactor model
Multi-factor
Expected returns
Cross section
Factors
Portfolio selection
Stochastic discount factor

All Science Journal Classification (ASJC) codes

  • Finance
  • Economics, Econometrics and Finance(all)

Cite this

Lozano Banda, Martín Carlos ; Hansen, Edwin ; Guidolin, Massimo. / Portfolio performance of linear SDF models : an out-of-sample assessment. In: Quantitative Finance. 2018 ; Vol. 18, No. 8. pp. 1425-1436.
@article{92ebc3fe47804dd7857d356e47435bf6,
title = "Portfolio performance of linear SDF models: an out-of-sample assessment",
abstract = "We evaluate linear stochastic discount factor models using an ex-post portfolio metric: the realized out-of-sample Sharpe ratio of mean–variance portfolios backed by alternative linear factor models. Using a sample of monthly US portfolio returns spanning the period 1968–2016, we find evidence that multifactor linear models have better empirical properties than the CAPM, not only when the cross-section of expected returns is evaluated in-sample, but also when they are used to inform one-month ahead portfolio selection. When we compare portfolios associated to multifactor models with mean–variance decisions implied by the single-factor CAPM, we document statistically significant differences in Sharpe ratios of up to 10 percent. Linear multifactor models that provide the best in-sample fit also yield the highest realized Sharpe ratios.",
author = "{Lozano Banda}, {Mart{\'i}n Carlos} and Edwin Hansen and Massimo Guidolin",
year = "2018",
month = "8",
day = "3",
doi = "10.1080/14697688.2018.1429646",
language = "English",
volume = "18",
pages = "1425--1436",
journal = "Quantitative Finance",
issn = "1469-7688",
publisher = "Routledge",
number = "8",

}

Portfolio performance of linear SDF models : an out-of-sample assessment. / Lozano Banda, Martín Carlos; Hansen, Edwin; Guidolin, Massimo.

In: Quantitative Finance, Vol. 18, No. 8, 03.08.2018, p. 1425-1436.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Portfolio performance of linear SDF models

T2 - an out-of-sample assessment

AU - Lozano Banda, Martín Carlos

AU - Hansen, Edwin

AU - Guidolin, Massimo

PY - 2018/8/3

Y1 - 2018/8/3

N2 - We evaluate linear stochastic discount factor models using an ex-post portfolio metric: the realized out-of-sample Sharpe ratio of mean–variance portfolios backed by alternative linear factor models. Using a sample of monthly US portfolio returns spanning the period 1968–2016, we find evidence that multifactor linear models have better empirical properties than the CAPM, not only when the cross-section of expected returns is evaluated in-sample, but also when they are used to inform one-month ahead portfolio selection. When we compare portfolios associated to multifactor models with mean–variance decisions implied by the single-factor CAPM, we document statistically significant differences in Sharpe ratios of up to 10 percent. Linear multifactor models that provide the best in-sample fit also yield the highest realized Sharpe ratios.

AB - We evaluate linear stochastic discount factor models using an ex-post portfolio metric: the realized out-of-sample Sharpe ratio of mean–variance portfolios backed by alternative linear factor models. Using a sample of monthly US portfolio returns spanning the period 1968–2016, we find evidence that multifactor linear models have better empirical properties than the CAPM, not only when the cross-section of expected returns is evaluated in-sample, but also when they are used to inform one-month ahead portfolio selection. When we compare portfolios associated to multifactor models with mean–variance decisions implied by the single-factor CAPM, we document statistically significant differences in Sharpe ratios of up to 10 percent. Linear multifactor models that provide the best in-sample fit also yield the highest realized Sharpe ratios.

UR - http://www.scopus.com/inward/record.url?scp=85042402020&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85042402020&partnerID=8YFLogxK

U2 - 10.1080/14697688.2018.1429646

DO - 10.1080/14697688.2018.1429646

M3 - Article

AN - SCOPUS:85042402020

VL - 18

SP - 1425

EP - 1436

JO - Quantitative Finance

JF - Quantitative Finance

SN - 1469-7688

IS - 8

ER -