Index tracking based on deep neural network

Abstract Deep learning has a strong ability to extract feature representations from data, since it has a great advantage in processing nonlinear and non-stationary data and reflecting nonlinear interactive relationship. This paper proposes to apply deep learning algorithms including deep neural network and deep autoencoder to track index performance and introduces a dynamic weight calculation method to measure the direct effects of the stocks on index. The empirical study takes historical data of Hang Seng Index (HSI) and its constituents to analyze the effectiveness and practicability of the index tracking method. The results show that the index tracking method based on deep neural network has a smaller tracking error, and thus can effectively track the index.

[1]  Nicholas G. Polson,et al.  Deep Portfolio Theory , 2016, ArXiv.

[2]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[3]  Craig Ellis,et al.  Can a neural network property portfolio selection process outperform the property market , 2005 .

[4]  Sergio Gómez,et al.  Portfolio selection using neural networks , 2005, Comput. Oper. Res..

[5]  H. Zimmermann,et al.  A linear model for tracking error minimization , 1999 .

[6]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[7]  Kin Keung Lai,et al.  Neural network-based mean-variance-skewness model for portfolio selection , 2008, Comput. Oper. Res..

[8]  Alberto Ferreira de Souza,et al.  Prediction-based portfolio optimization model using neural networks , 2009, Neurocomputing.

[9]  John E. Beasley,et al.  An evolutionary heuristic for the index tracking problem , 2003, Eur. J. Oper. Res..

[10]  R. Jansen,et al.  Optimal Benchmark Tracking with Small Portfolios , 2002 .

[11]  Tae Yoon Kim,et al.  Using genetic algorithm to support portfolio optimization for index fund management , 2005, Expert Syst. Appl..

[12]  W. Sharpe.  CAPITAL ASSET PRICES: A THEORY OF MARKET EQUILIBRIUM UNDER CONDITIONS OF RISK* , 1964 .

[13]  J. Lintner.  THE VALUATION OF RISK ASSETS AND THE SELECTION OF RISKY INVESTMENTS IN STOCK PORTFOLIOS AND CAPITAL BUDGETS , 1965 .

[14]  Alex Frino,et al.  Index Design and Implications for Index Tracking , 2004 .

[15]  Massimiliano Marcellino,et al.  Factor Based Index Tracking , 2002 .

[16]  Pascal Vincent,et al.  Unsupervised Feature Learning and Deep Learning: A Review and New Perspectives , 2012, ArXiv.

[17]  P. Kofman,et al.  Tracking Error and Active Portfolio Management , 2003 .

[18]  Hung Chun-Ping,et al.  An stock index replicating model based on time weighted SVM and it ’ s empirical analysis , 2014 .

[19]  Jan Hendrik Witte,et al.  Deep Learning for Finance: Deep Portfolios , 2016 .

[20]  R. Roll,et al.  A Mean/Variance Analysis of Tracking Error , 1992 .

[21]  Kees van Montfort,et al.  Index Tracking by Means of Optimized Sampling , 2008 .