Analysis of the five-factor asset pricing model with wavelet multiscaling approach

Anil K Bera, Umut Uyar, Sinem Guler Kangalli Uyar

Research output: Contribution to journalArticle

Abstract

We study the relationship between average returns and risk factors through wavelet multiscaling approach which enables us to investigate the risk-return relationship based on different time scales. The data for the period July 1963–February 2018 are gathered from the Kenneth French website. Each time series in the dataset is decomposed into five time scales. In order to make a comparison, the five-factor model is estimated based on both the scale basis and raw data. There are several key implications from our estimation results: i) The effects of risk factors on average returns vary over the time scales by their coefficient magnitudes and statistical significance. ii) Gibbons, Ross, and Shanken (1989) test results show that the intercepts of scale basis models are close to zero. iii) There is a period of unexpectedly higher cash flow for big value portfolios for long-term investments. iv) There is a minimum (maximum) risk level for aggressive (conservative) portfolios at different time horizons. Finally, we identify the risk factors in our five-factor model that have a significant effect on returns, and our model can capture the variations in average returns for different investment horizons.

Original languageEnglish (US)
JournalQuarterly Review of Economics and Finance
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

Factors
Asset pricing models
Wavelets
Risk factors
Time scales
Five-factor model
Time horizon
Risk-return
Cash flow
Web sites
Investment horizon
Make-to-order
Coefficients
Statistical significance

Keywords

  • Daubechies least asymmetric wavelet filter
  • Discrete wavelet transform
  • Investment horizon
  • Maximum overlap
  • The five-factor asset pricing model
  • Wavelet multiscaling approach

ASJC Scopus subject areas

  • Finance
  • Economics and Econometrics

Cite this

Analysis of the five-factor asset pricing model with wavelet multiscaling approach. / Bera, Anil K; Uyar, Umut; Kangalli Uyar, Sinem Guler.

In: Quarterly Review of Economics and Finance, 01.01.2019.

Research output: Contribution to journalArticle

@article{3cd8df4d64c347229151b79d3827c19a,
title = "Analysis of the five-factor asset pricing model with wavelet multiscaling approach",
abstract = "We study the relationship between average returns and risk factors through wavelet multiscaling approach which enables us to investigate the risk-return relationship based on different time scales. The data for the period July 1963–February 2018 are gathered from the Kenneth French website. Each time series in the dataset is decomposed into five time scales. In order to make a comparison, the five-factor model is estimated based on both the scale basis and raw data. There are several key implications from our estimation results: i) The effects of risk factors on average returns vary over the time scales by their coefficient magnitudes and statistical significance. ii) Gibbons, Ross, and Shanken (1989) test results show that the intercepts of scale basis models are close to zero. iii) There is a period of unexpectedly higher cash flow for big value portfolios for long-term investments. iv) There is a minimum (maximum) risk level for aggressive (conservative) portfolios at different time horizons. Finally, we identify the risk factors in our five-factor model that have a significant effect on returns, and our model can capture the variations in average returns for different investment horizons.",
keywords = "Daubechies least asymmetric wavelet filter, Discrete wavelet transform, Investment horizon, Maximum overlap, The five-factor asset pricing model, Wavelet multiscaling approach",
author = "Bera, {Anil K} and Umut Uyar and {Kangalli Uyar}, {Sinem Guler}",
year = "2019",
month = "1",
day = "1",
doi = "10.1016/j.qref.2019.09.014",
language = "English (US)",
journal = "Quarterly Review of Economics and Finance",
issn = "1062-9769",
publisher = "Elsevier",

}

TY - JOUR

T1 - Analysis of the five-factor asset pricing model with wavelet multiscaling approach

AU - Bera, Anil K

AU - Uyar, Umut

AU - Kangalli Uyar, Sinem Guler

PY - 2019/1/1

Y1 - 2019/1/1

N2 - We study the relationship between average returns and risk factors through wavelet multiscaling approach which enables us to investigate the risk-return relationship based on different time scales. The data for the period July 1963–February 2018 are gathered from the Kenneth French website. Each time series in the dataset is decomposed into five time scales. In order to make a comparison, the five-factor model is estimated based on both the scale basis and raw data. There are several key implications from our estimation results: i) The effects of risk factors on average returns vary over the time scales by their coefficient magnitudes and statistical significance. ii) Gibbons, Ross, and Shanken (1989) test results show that the intercepts of scale basis models are close to zero. iii) There is a period of unexpectedly higher cash flow for big value portfolios for long-term investments. iv) There is a minimum (maximum) risk level for aggressive (conservative) portfolios at different time horizons. Finally, we identify the risk factors in our five-factor model that have a significant effect on returns, and our model can capture the variations in average returns for different investment horizons.

AB - We study the relationship between average returns and risk factors through wavelet multiscaling approach which enables us to investigate the risk-return relationship based on different time scales. The data for the period July 1963–February 2018 are gathered from the Kenneth French website. Each time series in the dataset is decomposed into five time scales. In order to make a comparison, the five-factor model is estimated based on both the scale basis and raw data. There are several key implications from our estimation results: i) The effects of risk factors on average returns vary over the time scales by their coefficient magnitudes and statistical significance. ii) Gibbons, Ross, and Shanken (1989) test results show that the intercepts of scale basis models are close to zero. iii) There is a period of unexpectedly higher cash flow for big value portfolios for long-term investments. iv) There is a minimum (maximum) risk level for aggressive (conservative) portfolios at different time horizons. Finally, we identify the risk factors in our five-factor model that have a significant effect on returns, and our model can capture the variations in average returns for different investment horizons.

KW - Daubechies least asymmetric wavelet filter

KW - Discrete wavelet transform

KW - Investment horizon

KW - Maximum overlap

KW - The five-factor asset pricing model

KW - Wavelet multiscaling approach

UR - http://www.scopus.com/inward/record.url?scp=85072245368&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072245368&partnerID=8YFLogxK

U2 - 10.1016/j.qref.2019.09.014

DO - 10.1016/j.qref.2019.09.014

M3 - Article

AN - SCOPUS:85072245368

JO - Quarterly Review of Economics and Finance

JF - Quarterly Review of Economics and Finance

SN - 1062-9769

ER -