Conference Programme

Week 16-20 June

16 June

17 June

Pre Conference Workshop

Market Microstructure, Liquidity and Automated Trading

Pre Conference Workshop
Market Microstructure, Liquidity and Automated Trading
16—17 June, 2014

Session Titles:  Introduction to Market Microstructure and Liquidity Measures
Optimal Trade Execution Strategies
Automated Trading Strategies
Discussion of Trading Platforms and Their Features
Pre- and Post-Trade Analytics

Presenters Include:

  • Robert Kissell, Kissell Research Group
  • Dan diBartolomeo, Northfield Information Services
  • Ashok Banerjee, Finance Research/Trading Lab, Indian Institute of
  • Management
  • Rajib Ranjan Borah, iRage Capital Advisory Pvt. Ltd



18 June

19 June

Two Day Conference

Behavioural Models and Sentiment Analysis Applied to Finance

Session Titles:

Day 1:

  • Foundations & Technology for News Analytics
  • Text Mining and Sentiment Classification
  • Case Studies: (Fund Management, Trading Strategies, Risk Control)

Day 2:

  • Social Media, Micro Blogs, Google Trends
  • Other Asset Classes: Sentiment Analysis Applied to FX and Commodities
  • Case Studies: (Fund Management, Trading Strategies, Risk Control)

Presentation details received (in no particular order)

Quantifying Economic Behaviour Using Big Data
Tobias Preis, Warwick Business School and Artemis Capital Asset Management GmbH
In this talk, we will outline some recent highlights of our research, addressing two questions. Firstly, can big data resources provide insights into crises in financial markets? By analysing Google query volumes for search terms related to finance and views of Wikipedia articles, we find patterns which may be interpreted as early warning signs of stock market moves. Secondly, can we provide insight into international differences in economic wellbeing by comparing patterns of interaction with the Internet? To answer this question, we introduce a future-orientation index to quantify the degree to which Internet users seek more information about years in the future than years in the past. We analyse Google logs and find a striking correlation between the country's GDP and the predisposition of its inhabitants to look forward. Our results illustrate the potential that combining extensive behavioural data sets offers for a better understanding of large scale human economic behaviour.

Title: TBC
Stephen Pulman, Professor of Computational Linguistic Oxford University/TheySay (Analytics)

Sentiment, News and Volatility in both Stock Returns and Trading Volume
Dan diBartolomeo, Northfield Info Services Inc & CARISMA
Traditional views of financial markets operate with the conventional concept of time. Alternatively, one can readily explain the ebb and flow of trading volume and security returns through time via a construct in which the passage of time is perceived to speed up during period of heavy informational flows and perceived to slow down when information arrives only sparsely to market participants. We will use this alternative paradigm to examine the role of sentiment analysis, in which not only the amount of information arriving to markets changes over time, but also the character and apparent importance of this information flows. Our empirical evidence provides both agreement and contrast to previous related studies such as Kyle, Obizhaeva, Sinha and Tuzun (2012).

Title: TBC
Gautam Mitra, OptiRisk Systems & CARISMA

Title: TBC
James Cantarella, Thomson Reuters

Exploring Aggregate Sentiment and Investor Behavior
Elijah DePalma, Thomson Reuters

  • Impact of aggregate sentiment on market anomalies
  • Construction of aggregate sentiment signals using Thomson Reuters News Analytics
  • Exploring the value of aggregate sentiment beyond corporate news

Title: TBC
Peter Hafez, RavenPack

Title: TBC
Gary Kazantsev, Bloomberg

How to Use Language Recognition Systems to Improve Trading Models
Marco Dion,  Head of the Central Risk Book Trading Desk at JP Morgan

  • Company News flow has an undeniable impact on stock prices and can even represent the most important element defining a stock’s annual performance.
  • News items are fundamental and qualitative in nature, consequently academics and Quant houses have up to recently (i.e. pre 2007) struggled to investigate news flow from a systematic perspective.
  • This presentation shows how JP Morgan used the Thomson Reuters – News Analytics data to test various ideas and try to figure out if news flow can be used for longer-term investment (daily, weekly and monthly) by Quant Managers to generate alpha or manage risk.

Are you Trading Sentiment or Attention?
Raphael Markellos, Professor of Finance, Norwich Business School, University of East Anglia
Investor attention is a necessary but not sufficient condition for sentiment to have a tradable impact on markets. Sentiment and attention can be confused since they are related and have been proxied in a similar manner. I review similarities, differences, causes, effects and metrics for attention and sentiment. I also discuss their relative importance in the context of event-driven equity market trading strategies.

Sentiment Analysis for Commodity Markets
Svetlana Borovkova, Associate professor of Quantitative Finance, Vrije Universiteit Amsterdam

  • How commodity markets respond to positive and negative sentiment in news?
  • How do these responses vary for different commodities and states of the market?
  • What are the economic implications of price reactions to news?
  • Can we improve risk measurement and management tools by including news sentiment into the models?
  • Can we construct a continuous sentiment indicator for specific commodity and for the commodity market as a whole?

SOLID Project Results
Abby Levenberg, Senior Researcher at the Oxford-Man Institute of Quantitative Finance

  • Financial Prediction
  • Heterogeneous Big Data
  • Classifier Combination
  • Sentiment Analysis

Sentiment Analysis: Past, Present and Future
Bing Liu, Professor of Computer Science at the University of Illinois at Chicago
Sentiment analysis is the computational study of people’s opinions, sentiments, and emotions expressed in text. It has been studied extensively in natural language processing and Web data mining. It is also a core technology of social media analysis and has unlimited applications. In this talk, I will first define the problem and then discuss the current state-of-the-art and its future development.

What Does the Market Think? Evidence that Information Vividness and Trader Expectations Systematically Drive Currency, Commodity, and Global Equities Prices
Richard Peterson, MarketPsych

Researchers have identified that sentiment expressed in financial information can be predictive of market prices. Using the Thomson Reuters MarketPsych Indices, which comprise a multidimensional data feed derived from global news and social media content over 1998-2013, MarketPsych researchers decomposed the precise sentiments and themes that predictably drive asset prices over long-term horizons. Results indicate that specific political risks, macroeconomic themes, and sentiments influence collective investor behaviour, as seen in subsequent market price action with differences across asset classes. Uncertainty is most impactful to currencies, production volume is most significant for commodities, investor anger is the primary driver of U.S. equities, and global equity indexes are primarily influenced by government instability. This talk briefly touches on the psychological research supporting each identified effect including ambiguity aversion, emotional priming, and framing effects in individuals.

Other speakers agreed:

  • Elijah DePalma, Thomson Reuters
  • Peter Hafez, RavenPack
  • Gary Kazantsev, Bloomberg
  • Gautam Mitra, OptiRisk Systems & CARISMA
  • Stephen Pulman, Professor of Computational Linguistic Oxford University/TheySay (Analytics)




20 June

Post Conference Workshop

Sentiment Classification and Opinion Mining Using News Wires and Micro Blogs (Twitter)

Post Conference Workshop
Sentiment Classification and Opinion Mining Using News Wires and Micro Blogs (Twitter)
20 June, 2014

Title TBA
Stephen Pulman, Oxford University/TheySay Analytics

Aspect-based Sentiment Analysis
Bing Liu, Professor of Computer Science at the University of Illinois at Chicago
Analysis aims to extract opinions, sentiments and emotions from natural language text. Aspect-based analysis performs the fine-grained mining of opinions/sentiments and their targets. The targets are usually entities or entity aspects/attributes on which the opinions have been expressed. In this talk, I will first introduce this task, and then focus on its two core sub-tasks, aspect extraction and aspect sentiment classification, and present their representative algorithms.

Extracting User-Level Sentiments with Approval Relations. Enza Messina and Federico Alberto Pozzi, Department of Informatics Systems & Communication (DISCo) - University of Milano-Bicocca, Italy
In this talk we show how social relationships can be managed to improve user-level sentiment analysis of microblogs, overcoming the limitation of the state-of-the-art methods that generally consider posts as independent data. Early approaches consist in exploiting friendship relations, but since two friends could have different opinions about the same topic, it could however be inappropriate to measure sentiment similarity. To overcome these shortcomings, we present a framework that estimates user sentiments by combining post contents and approval relations, leading to significant improvements over the performance of complex classifiers based only on textual features.






Computer Weekly Logo

Computer Weekly Logo

CMMI Made Practical

SPIDER Network

Agile Holland