University of Leicester
Browse
1/1
2 files

Neural Network Copula Portfolio Optimization for Exchange Traded Funds

journal contribution
posted on 2017-11-17, 16:01 authored by Yang Zhao, Charalampos Stasinakis, Georgios Sermpinis, Yukun Shi
This paper attempts to investigate if adopting accurate forecasts from Neural Network (NN) models can lead to statistical and economically significant benefits in portfolio management decisions. In order to achieve that, three NNs, namely the Multi-Layer Perceptron (MLP), Recurrent Neural Network (RNN) and the Psi Sigma Network (PSN), are applied to the task of forecasting the daily returns of three Exchange Traded Funds (ETFs). The statistical and trading performance of the NNs is benchmarked with the traditional Autoregressive Moving Average (ARMA) models. Next, a novel dynamic asymmetric copula model (NNC) is introduced in order to capture the dependence structure across ETF returns. Based on the above, weekly re-balanced portfolios are obtained and compared by using the traditional mean-variance and the mean-CVaR portfolio optimization approach. In terms of the results, PSN outperforms all models in statistical and trading terms. Additionally, the asymmetric skewed t copula statistically outperforms symmetric copulas when it comes to modelling ETF returns dependence. The proposed NNC model leads to significant improvements in the portfolio optimization process, while forecasting covariance accounting for asymmetric dependence between the ETFs also improves the performance of obtained portfolios.

Funding

The work is partially supported by Project of Key Research Institute of Humanities & Social Science in Jiangxi Province [grant number JD16063].

History

Citation

Quantitative Finance, 2018

Author affiliation

/Organisation/COLLEGE OF SOCIAL SCIENCES, ARTS AND HUMANITIES/School of Management

Version

  • AM (Accepted Manuscript)

Published in

Quantitative Finance

Publisher

Taylor & Francis (Routledge)

issn

1469-7688

eissn

1469-7696

Acceptance date

2017-10-20

Copyright date

2017

Available date

2019-07-23

Publisher version

http://www.tandfonline.com/doi/abs/10.1080/14697688.2017.1414505

Notes

The file associated with this record is under embargo until 18 months after publication, in accordance with the publisher's self-archiving policy. The full text may be available through the publisher links provided above.

Language

en

Usage metrics

    University of Leicester Publications

    Categories

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC