Abstract: In this paper we investigate the profitability of a quantitative trading strategy based on Deep Learning methods. Specially we focus on a variant of the Recurrent Neural Network (RNN), the Long Short Term Memory Network (LSTM) and show its predictive power on stock price data. We use LSTM networks for selecting stocks using historical price.
The reason why RNNs are good for regression or classification of time series or data where time ordering matters is that RNNs capture the variation through time, thanks to its internal state dynamics. We made two studies, the rest focuses on predicting stock returns using one stock at a time. The hit-ratio in this experiment lies in the range 0.47 and 0.60 for the worst respectively best performing stock on unseen \live"" data.
The second experiment looks at the whole universe of stocks simultaneously. In this experiment our model achieves a hit-ratio between 0.50 and 0.71 on unseen \live"" data. From this experiment two portfolios were constructed, a long portfolio and a long-short portfolio with a Sharpe ratio of 8 respectively 10 for each of the portfolios. Our stock universe in both studies is composed of 50 stocks from the S&P 500.
Bio: Miquel Noguer i Alonso is a financial markets practitioner with more than 20 years of experience in asset management, he is currently Head of Development at Global AI ( Big Data Artificial Intelligence in Finance company ) and Head on Innovation and Technology at IEF. He worked for UBS AG (Switzerland) as Executive Director. He is member of European Investment Committee for the last 10 years. He worked as a Chief Investment Office and CIO for Andbank from 2000 to 2006. He started his career at KPMG.
He is Adjunct Professor at Columbia University teaching Asset Allocation, Big Data in Finance and Fintech. He is also Professor at ESADE teaching Hedge Fund, Big Data in Finance and Fintech. He taught the first Fintech and Big Data course at the London Business School in 2017.
He received an MBA and a Degree in business administration and economics in ESADE in 1993. In 2010 he earned a PhD in quantitative finance with a Summa Cum Laude distinction (UNED – Madrid Spain).
He completed a Postdoc in Columbia Business School in 2012. He collaborated with the Mathematics department of Fribourg during his PhD. He also holds the Certified European Financial Analyst (CEFA) 2000.
His research interests range from asset allocation, big data, machine learning to algorithmic trading and Fintech. His academic collaborations include a visiting scholarship in Columbia University in 2013 in the Finance and Economics Department, in Fribourg University in 2010 in the mathematics department, and giving presentations in Indiana University, ESADE and CAIA and several industry seminars like the Quant Summit USA 2017 and 2010.