Predicting Future Decisions with Deep learning for Financial Trading

Abstract: 

Markov Portfolio Theory has observed that investors put together a collection of stocks to build optimally diversified portfolios. The diversification is also emphasised in modern portfolio theory in neo-classical finance for rational investors. But despite the fact that such models provide effective normative explanations of how investors should behave, it is unknown whether investors ultimately do behave in this way.

This talk presents a Behavioural Finance research project focusing on the analysis of trader behaviour and whether it can be accurately emulated and predicted by analysing historical trader decision making using machine learning techniques. One aim is to try to imitate the behaviour (decision making) of traders based on their historical portfolio and market data, in order to predict their future decision making. The methods being used include inverse reinforcement learning and generative adversarial imitation learning, with a view to suggesting improvements and adjustments to adapt these models to be used with financial market data.

A second aim is to study visual trading patterns using candlestick pictures of stock data which are used by traders in their decision making. It is possible to apply machine learning techniques to extract useful patterns from the pictures and standard image classification methods have so far been applied, such as Convolutional Neural Networks (CNN) for multi financial time series image classification. However, since CNN does not have time-series features, Convolutional LSTM which is a combination of CNN and LSTM is being used, which allows for the capturing of time series data.

Bio: 

Ning Wang (Ph.D.) works as Senior Research Fellow in Data Science at the Oxford-NIE financial Big Data Lab, Mathematical Institute, University of Oxford. He also works as Research Associate at the Oxford Internet Institute.
His research is driven by a deep interest in analyzing a wide range of social and economic problems by exploiting data science approaches. His research interest lies in machine/deep learning in finance, social media, and mobile trading platform, online sentiment analysis and financial market, trading behaviors, and performance evaluation. He is also interested in the broad areas of behavioral finance, social computing, data mining, and social networks analysis.
Prior to Oxford, he was a postdoctoral researcher at the Computer Laboratory, University of Cambridge. He worked on the Social-based Forwarding Algorithm. He was involved in research to analyze a large number of real-world data on social networks in an effort to use centrality and community structure to optimize online communications.
He is also guest editor of a forthcoming Policy and Internet journal special issue of Chinese Web Data, program chair of the 14th IEEE Conference on Computer and Information Technology (CIT2014), Co-Chair of the 2014 IEEE Conference on Cyber, Physical and Social Computing (CPSCom2014), and TCP members of 2013 AAAI Conference on Weblogs and Social Media (ICWSM13) and reviewer for IEEE Communications Magazine.

Open Data Science

 

 

 

Open Data Science
One Broadway
Cambridge, MA 02142
info@odsc.com

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Youtube
Consent to display content from - Youtube
Vimeo
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google