Behavioural Models & Sentiment Analysis Applied to Finance

Behavioural Models & Sentiment Analysis Applied to Finance
Event on 2013-07-02 09:00:00

Sentiment analysis has developed as a technology that applies machine learning and makes a rapid assessment of the sentiments expressed in news releases. News events impact market sen-timent and financial news moves stock prices through a direct impact on a company’s expected future cash flows. This conference presents the current state of the art. It also explains how to ap-ply Sentiment Analysis to the respective models of trading, fund management and risk control. The conference will present in a summary form the research results in this fast-emerging field.



Preliminary Programme


Day  One: 2 July 2013


Text Mining for Systemic Risk

Sanjiv Das, Santa Clara University

Tools and techniques for text-mining to generate sentiment indexes, and assessment of predictive value


Network Finance Models

Sanjiv Das, Santa Clara University

Social network models in Finance. Illustrative applications to research in private equity, systemic risk. 


Trading Across Multiple Asset Classes Based on News Sentiment

Peter Hafez, RavenPack

Over the past few years, significant improvements have been made in the domain of sentiment analysis, and in understanding its impact on financial markets. Having primarily focused on equities, news analytics is now finding its way into other asset classes including FX and commodities. Peter Hafez  covers some of the latest research results of applying news analytics to finance. The presentation is backed up by a number of empirical studies covering (i) short-term Forex trading and (ii) longer-term stock selection


Panel 1: Adoption of sentiment analysis in fund management and trading strategies

Chairperson: Armando Gonzalez, RavenPack

Panellists include:

Peter van Kleef, Lakeview; Sanjiv Das, Santa Clara University; Gautam Mitra, OptiRisk

Other panellists to be invited


Inside the Global Brain: How sentiment trends in news and social media influence global equity and currency values

Richard Peterson, MarketPsych

"Achetez aux canon, vendez auz clarions" [Buy to the sound of cannons, sell to the sound of trumpets]

~ Attributed to Baron Nathan Rothschild, 1812.

Rothschild’s quote captures the essence of a contrarian country investment philosophy – the best time to invest in a nation is during a period of war or panic, and the best time to sell is during a time of contentment and celebration. The convergence of natural language processing, cloud computing, and map reduce architectures (big data), allow the quantification of real-time information flow into numerical streams useful in quantitative analysis. This presentation examines studies performed on a country-level sentiment data feed and concludes with a review of the impact of both country and currency sentiments on future global equity and currency pricing.


Overview of Sentiment Analysis applied to Finance

Gautam Mitra, OptiRisk Systems


New research results of sentiment analysis

Jacob Sisk / Aleksander Sobczyk, Thomson Reuters


Day  Two: 3 July 2013


Case Studies and Technology Overview


Case Study 1: TBA


Panel 2: Order Books, News Flows, News Metadata, Market Data: Predicting directional properties and volatility of asset prices

Panellists to be confirmed


Introducing Sentiment in the Predictive Analysis of Asset Behaviour: Return, Volatility and Liquidity in an Intra-day Setting

Xiang Yu, Researcher News Analytics; Keming Yu & Gautam Mitra, Brunel University.

We report an empirical study of a predictive analysis model for equities; the model uses high frequency (minute-bar) market data and quantified news sentiment data.  The purpose of the study is to identify a predictive model which can be used in designing automated trading strategies. Given that trading strategies take into consideration three important characteristics of an asset, namely, return, volatility and liquidity,  our model is designed to predict these three parameters for a collections of assets (finance industry stocks). The minute-bar market data as well as intraday news sentiment metadata have been provided by Thomson Reuters. 


Case Study 2: TBA


Panel 3: Social media meta data, trust, information leakage: impact on trading and compliance

Panellists to be confirmed


Case Study 3: TBA


Case Study 4: TBA


at Fitch 7 City Learning
4 Chiswell Street
City of London, United Kingdom

This entry was posted in Natural Language and tagged , , , , , . Bookmark the permalink.

Leave a Reply