Financial sentiment analysis is a growing area of interest for major financial institutions such as banks and hedge funds. Automated trading strategies based on traditional market analysis are producing lower returns as institutions compete with similar strategies. Automated sentiment analysis offers a completely new data stream that can break this deadlock. Companies who get it right will have a significant advantage over their competition.
Bitext attended the Behavioural Models & Sentiment Analysis Applied to Finance conference in London on 2nd & 3rd July. A large audience of quantitative analysts watched presentations from academics, vendors and trading strategists. The emphasis was on proven strategies. Backtesting on market data showed the value of automated sentiment analysis in a variety of different trading strategies. One point was repeated by the speakers: high-quality text analytics is crucial to the success of these strategies; detailed and accurate entity extraction, sentiment analysis and categorization make the difference between profits and losses. The clear message was: the financial community have the expertise in trading, what they need is the best possible linguistic analysis to support them.
An exciting topic of discussion was the role that sentiment analysis of Social Media can play in the fundamental analysis of companies. For institutions that base their strategies on company earnings, the ability to predict whether a company will meet its targets is critical. Social Media sentiment analysis can automate this, re-using techniques pioneered in marketing by companies such as Salesforce/Radian6, who use Bitext's multilingual technology. Predictive analysis based on what consumers are saying makes it possible to see whether sales will match forecasts. Knowing this in advance of any announcement is a clear competitive advantage.