Policy, Sentiment, and Financial Markets

05 June 2015

Policy, Sentiment, and Financial Markets (POLSENT) is a DCU-based Irish Research Council-funded interdisciplinary research collaboration between the School of Law and Government, the School of Communications and the Insight Centre for Data Analytics.

Led by School of Law and Government researcher and lecturer Dr Michael Breen, POLSENT examines the role of the media in affecting the financial markets. Dr Breen is working in conjunction with Prof Iain McMenamin, Dr Gemma McNulty and Dr Michael Courtney on this project.

Working with an economic theory called the Efficient Market Hypothesis (EMH), this research project is exploring whether the market simply responds to new information (via media coverage) or whether news stories in general, about political competition or political personalities, influence investor behaviour. POLSENT is also looking at several other attributes of the news, particularly news about macroeconomic policy. The idea is to study not only news that says ‘the economy is doing good or bad today’ but news about actual policy change. From this perspective, actual policy e.g. a change in interest rates, rather than sentiment, can be classified.

The POLSENT group has collected millions of newspaper articles with the aim of developing ways of classifying the news in order to understand what political news is. Dr Breen and his team have been collaborating with Insight, using new methods from computer science, including machine learning, to automatically classify news stories gathered from selected European print media. The first phase of the project involves the manual coding and classification of thousands of newspaper articles. These articles are then used as training material for a “computer brain” that learns to spot the same coding patterns in a process known as supervised learning.

Michael Breen, Iain McMenamin, Michael Courtney and Gemma McNulty. 2015. Daily Judgement: Political News and Financial Markets. POLSENT WP No. 1. Dublin City University.
Michael Breen, Iain McMenamin, Michael Courtney and Gemma McNulty. 2015. A Generalisable Approach to Machine Learning and Media Content. POLSENT WP No. 2. Dublin City University.

Related News

This website uses cookies to ensure you get the best experience. The majority of the cookies used on this website are associated with analytics, collecting information about how visitors use our site. The cookies collect information in an anonymous form that does not identify an individual. Learn more
Current status: AcceptedDeclinedNot yet accepted