Out-of-sample Performance-based Estimation of Expected Returns for Portfolio Selection

50 Pages Posted: 26 Aug 2022

See all articles by Peng-Chu Chen

Peng-Chu Chen

The University of Hong Kong

Yan Wang

The University of Hong Kong

Date Written: August 19, 2022

Abstract

This paper provides a framework for obtaining the estimator of expected asset returns for portfolio selection. The framework relies on a linear model where the expected returns are the coefficients to be estimated. The model is fitted to a synthetic dataset by Bayesian regression. The estimator is computed using a Gibbs sampler; it is consistent and asymptotically efficient when the size of the synthetic dataset grows to infinity. An empirical study shows that, under appropriate conditions, mean-variance portfolios constructed using this estimator yield better out-of-sample average returns and Sharpe ratios than benchmark portfolios, with or without a norm constraint.

Keywords: portfolio selection, expected return estimation, estimation error, Bayes, Gibbs sampler, data augmentation, synthetic data

JEL Classification: C10, G11

Suggested Citation

Chen, Peng-Chu and Wang, Yan, Out-of-sample Performance-based Estimation of Expected Returns for Portfolio Selection (August 19, 2022). Available at SSRN: https://ssrn.com/abstract=4194264 or http://dx.doi.org/10.2139/ssrn.4194264

Peng-Chu Chen (Contact Author)

The University of Hong Kong ( email )

Pokfulam Road
Hong Kong, Pokfulam HK
China

Yan Wang

The University of Hong Kong ( email )

Pokfulam Road
Hong Kong, Pokfulam HK
China

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

Paper statistics

Downloads
29
Abstract Views
379
PlumX Metrics