Machine learning, anomalies, and the expected market return: Evidence from China

Qingjie Du; Yang Wang; Chi shen Wei; K. C.John Wei

doi:10.1016/j.pacfin.2023.102168

Machine learning, anomalies, and the expected market return: Evidence from China

Qingjie Du, Yang Wang, Chi shen Wei, K. C.John Wei

Research output: Journal article publication › Journal article › Academic research › peer-review

Abstract

We investigate whether machine learning (ML) techniques that forecast overall U.S. market returns using cross-sectional stock return anomalies in Dong et al. (2022) are useful for the China equity market. We successfully forecast out-of-sample R² of the market return in China using a combined version of ordinary least squares and an elastic net model. However, the other four ML methods cannot forecast the market return. Overall, our exercise highlights the potential of ML techniques, but also calls for future research to rule out the possibility of model mining.

Original language	English
Article number	102168
Journal	Pacific Basin Finance Journal
Volume	82
DOIs	https://doi.org/10.1016/j.pacfin.2023.102168
Publication status	Published - Dec 2023

Keywords

Anomalies
Chinese stock market
Machine learning
Return predictability

ASJC Scopus subject areas

Finance
Economics and Econometrics

More information

10.1016/j.pacfin.2023.102168

Cite this

@article{b7bc92b1a6854a03ad48021eb4677c1c,

title = "Machine learning, anomalies, and the expected market return: Evidence from China",

abstract = "We investigate whether machine learning (ML) techniques that forecast overall U.S. market returns using cross-sectional stock return anomalies in Dong et al. (2022) are useful for the China equity market. We successfully forecast out-of-sample R2 of the market return in China using a combined version of ordinary least squares and an elastic net model. However, the other four ML methods cannot forecast the market return. Overall, our exercise highlights the potential of ML techniques, but also calls for future research to rule out the possibility of model mining.",

keywords = "Anomalies, Chinese stock market, Machine learning, Return predictability",

author = "Qingjie Du and Yang Wang and Wei, {Chi shen} and Wei, {K. C.John}",

note = "Publisher Copyright: {\textcopyright} 2023",

year = "2023",

month = dec,

doi = "10.1016/j.pacfin.2023.102168",

language = "English",

volume = "82",

journal = "Pacific Basin Finance Journal",

issn = "0927-538X",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Machine learning, anomalies, and the expected market return

T2 - Evidence from China

AU - Du, Qingjie

AU - Wang, Yang

AU - Wei, Chi shen

AU - Wei, K. C.John

PY - 2023/12

Y1 - 2023/12

N2 - We investigate whether machine learning (ML) techniques that forecast overall U.S. market returns using cross-sectional stock return anomalies in Dong et al. (2022) are useful for the China equity market. We successfully forecast out-of-sample R2 of the market return in China using a combined version of ordinary least squares and an elastic net model. However, the other four ML methods cannot forecast the market return. Overall, our exercise highlights the potential of ML techniques, but also calls for future research to rule out the possibility of model mining.

AB - We investigate whether machine learning (ML) techniques that forecast overall U.S. market returns using cross-sectional stock return anomalies in Dong et al. (2022) are useful for the China equity market. We successfully forecast out-of-sample R2 of the market return in China using a combined version of ordinary least squares and an elastic net model. However, the other four ML methods cannot forecast the market return. Overall, our exercise highlights the potential of ML techniques, but also calls for future research to rule out the possibility of model mining.

KW - Anomalies

KW - Chinese stock market

KW - Machine learning

KW - Return predictability

UR - http://www.scopus.com/inward/record.url?scp=85173300346&partnerID=8YFLogxK

U2 - 10.1016/j.pacfin.2023.102168

DO - 10.1016/j.pacfin.2023.102168

M3 - Journal article

AN - SCOPUS:85173300346

SN - 0927-538X

VL - 82

JO - Pacific Basin Finance Journal

JF - Pacific Basin Finance Journal

M1 - 102168

ER -

Machine learning, anomalies, and the expected market return: Evidence from China

Abstract

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this