Fama–French Five-Factor Modeling: New Evidence from a Nonparametric Method

Zihao Hou; Viktor Manahov; Dimitrios Stafylas

doi:10.1515/snde-2024-0046

Fama–French Five-Factor Modeling: New Evidence from a Nonparametric Method

Zihao Hou^*, Viktor Manahov, Dimitrios Stafylas

^*Corresponding author for this work

The York Management School

Research output: Contribution to journal › Article › peer-review

Abstract

This study constructs a Fama-French five-factor model that considers the time-varying properties of the parameters and introduces a nonparametric method that estimates the factor loadings. We approach the topic from a micro perspective using high-frequency data to construct factors and models to evaluate the sensitivity of each factor on abnormal returns. The results show that the conditional alphas of portfolios are optimised, and the nonparametric model outperforms the traditional models. Our findings lead investors to consider the impact of parameter time-variation when using multi-factor stock selection models to construct asset portfolios.

Original language	English
Journal	Studies in Nonlinear Dynamics & Econometrics
Volume	30
Issue number	1
DOIs	https://doi.org/10.1515/snde-2024-0046
Publication status	Published - 11 Mar 2025

Access to Document

10.1515/snde-2024-0046Licence: CC BY

Cite this

@article{6f88715488aa4a8f935bac04891ef415,

title = "Fama–French Five-Factor Modeling: New Evidence from a Nonparametric Method",

abstract = "This study constructs a Fama-French five-factor model that considers the time-varying properties of the parameters and introduces a nonparametric method that estimates the factor loadings. We approach the topic from a micro perspective using high-frequency data to construct factors and models to evaluate the sensitivity of each factor on abnormal returns. The results show that the conditional alphas of portfolios are optimised, and the nonparametric model outperforms the traditional models. Our findings lead investors to consider the impact of parameter time-variation when using multi-factor stock selection models to construct asset portfolios.",

author = "Zihao Hou and Viktor Manahov and Dimitrios Stafylas",

year = "2025",

month = mar,

day = "11",

doi = "10.1515/snde-2024-0046",

language = "English",

volume = "30",

journal = "Studies in Nonlinear Dynamics & Econometrics",

issn = "1558-3708",

publisher = "Berkeley Electronic Press",

number = "1",

}

TY - JOUR

T1 - Fama–French Five-Factor Modeling

T2 - New Evidence from a Nonparametric Method

AU - Hou, Zihao

AU - Manahov, Viktor

AU - Stafylas, Dimitrios

PY - 2025/3/11

Y1 - 2025/3/11

N2 - This study constructs a Fama-French five-factor model that considers the time-varying properties of the parameters and introduces a nonparametric method that estimates the factor loadings. We approach the topic from a micro perspective using high-frequency data to construct factors and models to evaluate the sensitivity of each factor on abnormal returns. The results show that the conditional alphas of portfolios are optimised, and the nonparametric model outperforms the traditional models. Our findings lead investors to consider the impact of parameter time-variation when using multi-factor stock selection models to construct asset portfolios.

AB - This study constructs a Fama-French five-factor model that considers the time-varying properties of the parameters and introduces a nonparametric method that estimates the factor loadings. We approach the topic from a micro perspective using high-frequency data to construct factors and models to evaluate the sensitivity of each factor on abnormal returns. The results show that the conditional alphas of portfolios are optimised, and the nonparametric model outperforms the traditional models. Our findings lead investors to consider the impact of parameter time-variation when using multi-factor stock selection models to construct asset portfolios.

U2 - 10.1515/snde-2024-0046

DO - 10.1515/snde-2024-0046

M3 - Article

SN - 1558-3708

VL - 30

JO - Studies in Nonlinear Dynamics & Econometrics

JF - Studies in Nonlinear Dynamics & Econometrics

IS - 1

ER -