Online Program

Return to main conference page
Saturday, May 19
Applications
Scientific and Financial Modeling
Sat, May 19, 10:30 AM - 12:00 PM
Lake Fairfax A
 

Data Driven Portfolio Optimization Utilizing Machine Learning (304623)

Presentation

*Melinda Hsieh, Rider University 

Keywords: data-driven, optimal portfolio, machine learning,

In practice, data-driven optimal portfolio decisions are derived based on the time series data of observed target asset price. Such data-driven optimization rules are prone to have inferior out-of-sample performance due to estimation errors of parameters plugged in the optimization setting. In the 'big data' era, correlations between the target asset prices and other axillary variables are frequently observed. These auxiliary variables, if used wisely, can provide valuable information on their association with the target asset prices and thus have the potential to improve the out-of-sample performance of the formed optimal portfolio. In this paper, several machine learning methods are applied to leverage the association between target and auxiliary variables and to derive the optimal portfolio decisions. A comparison study on the out-of-sample performance of the formed portfolio with and without utilizing machine learning methods is conducted.