Armando Gonzalez
To make business decisions, people have been manually extracting patterns from news
for centuries, but the increasing volume of news in modern times has called for more
automated approaches. News analysis can now be done by collecting and processing
information using computer systems equipped to handle the intricacies of the human
language.
A vast amount of textual information is available in real time through newswire
services and even more is updated on a regular basis on the web. Newswires are delivered
by news agencies that provide information from around the world, covering breaking
events in business, finance, politics, entertainment, and more. Professional newswire
services are organizations of journalists established to supply news reports to other
organizations in the news trade: newspapers, magazines, and radio and television broad-
casters. The major news agencies generally prepare hard-news stories and feature articles
that can be used by other organizations with little or no modification, and then sell them
to other news providers.
Today, computers can quantify many aspects of textual content including positive
and negative perceptions on facts and opinions reported in the news. They can con-
tinuously analyze relevant information from all major news and internet sources like
financial sites, blogs, and local and regional newspapers to produce real-time news
sentiment scores. These scores not only allow market participants to capture market
opportunities, they also help to improve risk management and provide for better trading
execution. Financial firms use these data to lead portfolio allocations, which help
improve the average life and profit of existing trading models. They can enhance risk
metrics such as Value at Risk calculations and control ramp-up or ramp-down times in
algorithmic trading engines.
Many high-frequency traders currently boost their gains using news-based algorithms
to speed their response time to breaking events. They build defensive applications to
ensure that key news events are factored in, and use automated news analysis as a more
effective form of low-latency decision support. Firms also reduce the time required for
low-frequency fundamental traders to assess their options manually and execute their
responses more effectively. Structured news can also help in post-trade analysis to
explain why a traditional algorithm or strategy did not work.
Another benefit from automated trading on news is that it can protect portfolio
managers or traders from the consequences of missing important news that has an
The Handbook of News Analytics in Finance Edited by L. Mitra and G. Mitra
#2011 John Wiley & Sons