Predicting Returns Out of Sample: A Naïve Model Averaging Approach

101 Pages Posted: 27 Sep 2019 Last revised: 17 Oct 2022

See all articles by Huafeng (Jason) Chen

Huafeng (Jason) Chen

Fudan University - Fanhai International School of Finance (FISF)

Liang Jiang

Fudan University - Fanhai International School of Finance (FISF)

Weiwei Liu

Tsinghua University - PBC School of Finance (PBCSF)

Date Written: October 12, 2022

Abstract

We propose a naïve model averaging (NMA) method, averaging the OLS out-of-sample forecasts and the historical means, that produces mostly positive out-of-sample R2s for the variables that are significant in sample in forecasting market returns. Surprisingly, more sophisticated weighting schemes that combine the predictive variable and historical mean do not consistently perform better. With unstable economic relations and a limited sample size, sophisticated methods may lead to overfitting or be subject to more estimation errors. In such situations, our simple methods may work better. Model misspecification, rather than declining return predictability, likely explains the predictive performance of the NMA method.

Keywords: out of sample tests, naïve model averaging, market returns, ridge regression

JEL Classification: G12, G11

Suggested Citation

Chen, Huafeng (Jason) and Jiang, Liang and Liu, Weiwei, Predicting Returns Out of Sample: A Naïve Model Averaging Approach (October 12, 2022). Available at SSRN: https://ssrn.com/abstract=3455866 or http://dx.doi.org/10.2139/ssrn.3455866

Huafeng (Jason) Chen

Fudan University - Fanhai International School of Finance (FISF) ( email )

220 Handan Road
Shanghai, 200433
China

Liang Jiang

Fudan University - Fanhai International School of Finance (FISF) ( email )

220 Handan Road
Shanghai, 200433
China

Weiwei Liu (Contact Author)

Tsinghua University - PBC School of Finance (PBCSF) ( email )

Beijing, Haidian Distreet, Chengfu Road NO.43
Beijing, 100083
China
15600325656 (Phone)

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
133
Abstract Views
953
Rank
341,614
PlumX Metrics