Revisiting Fama–French factors' predictability with Bayesian modelling and copula‐based portfolio optimization

Yang Zhao, Charalampos Stasinakis (Corresponding Author), Georgios Sermpinis, Filipa Da Silva Fernandes

Research output: Contribution to journalArticle

Abstract

This study is investigating the predictability of the five Fama-French factors and explores their optimal portfolio allocation for factor investing during 2000-2017. Firstly, we forecast each factor with a pool of linear and non-linear models. Next, the individual forecasts are combined through Dynamic Model Averaging (DMA), while their performance is benchmarked by the best performing individual predictor and other forecast combination techniques. Finally, we use the Generalized Autoregressive Score (GAS) model and the skewed t copula method to estimate the correlation of assets. The GAS performance is also compared with other traditional approaches such as Dynamic Conditional Correlation (DCC) model and Asymmetric Dynamic Conditional Correlation (ADCC). The performance of the constructed portfolios is assessed through traditional metrics and ratios accounting for the Conditional Value-at-Risk (CVaR) and the Conditional Diversification Benefits (CDB) approach. Our results show that combining Bayesian forecast combinations with copulas is leading to significant improvements in the portfolio optimization process, while forecasting covariance accounting for asymmetric dependence between the factors adds diversification benefits to the obtained portfolios.
Original languageEnglish
JournalInternational Journal of Finance and Economics
Early online date5 Aug 2019
DOIs
Publication statusE-pub ahead of print - 5 Aug 2019

Fingerprint

Portfolio optimization
Factors
Bayesian modeling
Predictability
Copula
Diversification benefits
Dynamic conditional correlation
Forecast combination
Asymmetric dependence
Investing
Portfolio allocation
Conditional value at risk
Assets
Model averaging
Optimal portfolio
Predictors

Keywords

  • Factor Investing
  • Portfolio Optimization
  • Dynamic Model Averaging
  • Forecast Combinations
  • forecast combinations
  • dynamic model averaging
  • factor investing
  • portfolio optimization
  • support vector regression
  • returns
  • risk
  • inflation
  • dynamics
  • krill herd
  • SUPPORT VECTOR REGRESSION
  • RETURNS
  • RISK
  • INFLATION
  • DYNAMICS
  • KRILL HERD

ASJC Scopus subject areas

  • Economics and Econometrics
  • Accounting
  • Finance

Cite this

Revisiting Fama–French factors' predictability with Bayesian modelling and copula‐based portfolio optimization. / Zhao, Yang; Stasinakis, Charalampos (Corresponding Author); Sermpinis, Georgios; Fernandes, Filipa Da Silva.

In: International Journal of Finance and Economics, 05.08.2019.

Research output: Contribution to journalArticle

@article{a1f2f69e63884f3d8148e202dc980ffe,
title = "Revisiting Fama–French factors' predictability with Bayesian modelling and copula‐based portfolio optimization",
abstract = "This study is investigating the predictability of the five Fama-French factors and explores their optimal portfolio allocation for factor investing during 2000-2017. Firstly, we forecast each factor with a pool of linear and non-linear models. Next, the individual forecasts are combined through Dynamic Model Averaging (DMA), while their performance is benchmarked by the best performing individual predictor and other forecast combination techniques. Finally, we use the Generalized Autoregressive Score (GAS) model and the skewed t copula method to estimate the correlation of assets. The GAS performance is also compared with other traditional approaches such as Dynamic Conditional Correlation (DCC) model and Asymmetric Dynamic Conditional Correlation (ADCC). The performance of the constructed portfolios is assessed through traditional metrics and ratios accounting for the Conditional Value-at-Risk (CVaR) and the Conditional Diversification Benefits (CDB) approach. Our results show that combining Bayesian forecast combinations with copulas is leading to significant improvements in the portfolio optimization process, while forecasting covariance accounting for asymmetric dependence between the factors adds diversification benefits to the obtained portfolios.",
keywords = "Factor Investing, Portfolio Optimization, Dynamic Model Averaging, Forecast Combinations, forecast combinations, dynamic model averaging, factor investing, portfolio optimization, support vector regression, returns, risk, inflation, dynamics, krill herd, SUPPORT VECTOR REGRESSION, RETURNS, RISK, INFLATION, DYNAMICS, KRILL HERD",
author = "Yang Zhao and Charalampos Stasinakis and Georgios Sermpinis and Fernandes, {Filipa Da Silva}",
year = "2019",
month = "8",
day = "5",
doi = "10.1002/ijfe.1742",
language = "English",
journal = "International Journal of Finance and Economics",
issn = "1076-9307",
publisher = "John Wiley & Sons Ltd.",

}

TY - JOUR

T1 - Revisiting Fama–French factors' predictability with Bayesian modelling and copula‐based portfolio optimization

AU - Zhao, Yang

AU - Stasinakis, Charalampos

AU - Sermpinis, Georgios

AU - Fernandes, Filipa Da Silva

PY - 2019/8/5

Y1 - 2019/8/5

N2 - This study is investigating the predictability of the five Fama-French factors and explores their optimal portfolio allocation for factor investing during 2000-2017. Firstly, we forecast each factor with a pool of linear and non-linear models. Next, the individual forecasts are combined through Dynamic Model Averaging (DMA), while their performance is benchmarked by the best performing individual predictor and other forecast combination techniques. Finally, we use the Generalized Autoregressive Score (GAS) model and the skewed t copula method to estimate the correlation of assets. The GAS performance is also compared with other traditional approaches such as Dynamic Conditional Correlation (DCC) model and Asymmetric Dynamic Conditional Correlation (ADCC). The performance of the constructed portfolios is assessed through traditional metrics and ratios accounting for the Conditional Value-at-Risk (CVaR) and the Conditional Diversification Benefits (CDB) approach. Our results show that combining Bayesian forecast combinations with copulas is leading to significant improvements in the portfolio optimization process, while forecasting covariance accounting for asymmetric dependence between the factors adds diversification benefits to the obtained portfolios.

AB - This study is investigating the predictability of the five Fama-French factors and explores their optimal portfolio allocation for factor investing during 2000-2017. Firstly, we forecast each factor with a pool of linear and non-linear models. Next, the individual forecasts are combined through Dynamic Model Averaging (DMA), while their performance is benchmarked by the best performing individual predictor and other forecast combination techniques. Finally, we use the Generalized Autoregressive Score (GAS) model and the skewed t copula method to estimate the correlation of assets. The GAS performance is also compared with other traditional approaches such as Dynamic Conditional Correlation (DCC) model and Asymmetric Dynamic Conditional Correlation (ADCC). The performance of the constructed portfolios is assessed through traditional metrics and ratios accounting for the Conditional Value-at-Risk (CVaR) and the Conditional Diversification Benefits (CDB) approach. Our results show that combining Bayesian forecast combinations with copulas is leading to significant improvements in the portfolio optimization process, while forecasting covariance accounting for asymmetric dependence between the factors adds diversification benefits to the obtained portfolios.

KW - Factor Investing

KW - Portfolio Optimization

KW - Dynamic Model Averaging

KW - Forecast Combinations

KW - forecast combinations

KW - dynamic model averaging

KW - factor investing

KW - portfolio optimization

KW - support vector regression

KW - returns

KW - risk

KW - inflation

KW - dynamics

KW - krill herd

KW - SUPPORT VECTOR REGRESSION

KW - RETURNS

KW - RISK

KW - INFLATION

KW - DYNAMICS

KW - KRILL HERD

UR - http://www.scopus.com/inward/record.url?scp=85070667915&partnerID=8YFLogxK

UR - http://www.mendeley.com/research/revisiting-famafrench-factors-predictability-bayesian-modelling-copulabased-portfolio-optimization

UR - https://abdn.pure.elsevier.com/en/en/researchoutput/revisiting-famafrench-factors-predictability-with-bayesian-modelling-and-copulabased-portfolio-optimization(a1f2f69e-6388-4f3d-8148-e202dc980ffe).html

U2 - 10.1002/ijfe.1742

DO - 10.1002/ijfe.1742

M3 - Article

JO - International Journal of Finance and Economics

JF - International Journal of Finance and Economics

SN - 1076-9307

ER -