Optimal Asset Allocation for Outperforming a Stochastic Benchmark Target

33 Pages Posted: 29 Jun 2020

See all articles by Chendi Ni

Chendi Ni

University of Waterloo - David R Cheriton School of Computer Science

Yuying Li

University of Waterloo

Peter Forsyth

University of Waterloo - David R. Cheriton School of Computer Science

Ray Carroll

Neuberger Berman Breton Hill

Date Written: June 4, 2020

Abstract

We propose a data-driven Neural Network (NN) optimization framework to determine the optimal multi-period dynamic asset allocation strategy for outperforming a general stochastic target. We formulate the problem as an optimal stochastic control with an asymmetric, distribution shaping, objective function. The proposed framework is illustrated with the asset allocation problem in the accumulation phase of a defined contribution pension plan, with the goal of achieving a higher terminal wealth than a stochastic benchmark. We demonstrate that the data-driven approach is capable of learning an adaptive asset allocation strategy directly from historical market returns, without assuming any parametric model of the financial market dynamics. Following the optimal adaptive strategy, investors can make allocation decisions simply depending on the current state of the portfolio. The optimal adaptive strategy outperforms the benchmark constant proportion strategy, achieving a higher terminal wealth with a 90% probability, a 46% higher median terminal wealth, and a significantly more right-skewed terminal wealth distribution. We further demonstrate the robustness of the optimal adaptive strategy by testing the performance of the strategy on bootstrap resampled market data, which has different distributions compared to the training data.

Keywords: stochastic benchmark, portfolio allocation, neural network

JEL Classification: G11, C45

Suggested Citation

Ni, Chendi and Li, Yuying and Forsyth, Peter and Carroll, Ray, Optimal Asset Allocation for Outperforming a Stochastic Benchmark Target (June 4, 2020). Available at SSRN: https://ssrn.com/abstract=3619332 or http://dx.doi.org/10.2139/ssrn.3619332

Chendi Ni

University of Waterloo - David R Cheriton School of Computer Science

Waterloo, On N2l 3G1
Canada

Yuying Li

University of Waterloo ( email )

Waterloo, Ontario N2L 3G1
Canada

Peter Forsyth (Contact Author)

University of Waterloo - David R. Cheriton School of Computer Science ( email )

200 University Avenue West
Waterloo, ON
Canada

Ray Carroll

Neuberger Berman Breton Hill ( email )

2 Bloor St East
Toronto, Ontario M4W 1A8
Canada

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

Paper statistics

Downloads
53
Abstract Views
301
rank
443,991
PlumX Metrics