A Comparative Study of Traditional Linear Models and Nonlinear Neural Network Model on Asset Pricing

Wen Wang

2022

Abstract

Models in traditional asset pricing theories, such as CAPM and the Fama-French three-factor model, explain the linear relationship between market returns, company size, company type, and return on assets. But in a more complex financial market, the linear relationship contained in the above model may not hold. Therefore, the main focus of this paper is to analyze the nonlinearity between stock excess return and its influencing factors. The existence of nonlinearity is confirmed via the RESET test proposed by Ramsey. Then, the nonlinear neural network model is used to further study the nonlinear relationship. Based on the data of the A-share market, it is verified that there is a nonlinear relationship between stock excess returns and their influencing factors, and the nonlinear neural network model shows better prediction performance than traditional linear models.

Paper Citation

in Harvard Style

Wang W. (2022). A Comparative Study of Traditional Linear Models and Nonlinear Neural Network Model on Asset Pricing. In Proceedings of the 4th International Conference on Economic Management and Model Engineering - Volume 1: ICEMME; ISBN 978-989-758-636-1, SciTePress, pages 535-541. DOI: 10.5220/0012036400003620

in Bibtex Style

@conference{icemme22,
author={Wen Wang},
title={A Comparative Study of Traditional Linear Models and Nonlinear Neural Network Model on Asset Pricing},
booktitle={Proceedings of the 4th International Conference on Economic Management and Model Engineering - Volume 1: ICEMME},
year={2022},
pages={535-541},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012036400003620},
isbn={978-989-758-636-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 4th International Conference on Economic Management and Model Engineering - Volume 1: ICEMME
TI - A Comparative Study of Traditional Linear Models and Nonlinear Neural Network Model on Asset Pricing
SN - 978-989-758-636-1
AU - Wang W.
PY - 2022
SP - 535
EP - 541
DO - 10.5220/0012036400003620
PB - SciTePress