The Handbook of News Analytics in Finance

14

Incorporating news into algorithmic trading strategies: Increasing the signal-to-noise ratio

Richard Brown

While the mass adoption of machine-readable news in trading environments is still in its early stages, there are a number of techniques that can be used to significantly improve trading performance, both in offensive as well as in defensive strategies. These techniques range from simple circuit breakers or wolf detection systems (defensive) to systems that will exploit the volatility surrounding significant news items, or those that will predict the direction and magnitude of a price movement (offensive).

Over the last few years, there has been a substantially larger amount of research on the use of news for trading and investment strategies. The research has been less conclusive on signals for returns, however. This has in large part been due to two factors. First, there has traditionally been a lack of comprehensive metrics and metadata, which can be used to determine the direction, magnitude, and duration of such movements. Second, logic suggests that if there is a killer strategy for doing this, one is more likely to trade on it than to tell the whole world about it.

So, how can one incorporate news into algorithmic strategies to improve trading performance?

One of the more common and easily implemented uses of machine-readable news in a trading environment is to use the news feed as a circuit breaker for algorithms. This approach would stop your algorithm from its ...