Sentiment in financial news articles drives change in asset prices, volatility and volume. How can investors use sentiment data and news analytics to decipher the longer-term direction of price movements and make informed and profitable investment decisions?
- When analyzing market anomalies, incorporating news analytics and sentiment data can enable investors to gain an advantage, and make informed and profitable decisions.
- Refinitiv News Analytics can be used by investors to profit from large earnings surprises, where sentiment data is used to gauge the excess profits that could be made from these market anomalies.
- News sentiment data can also be used in pricing risk. Refinitiv News Analytics can be used to scrutinize how sentiment data affects the capital asset pricing model.
News sentiment data can help investors to gain an advantage from market anomalies such as post-earnings announcement drift and pricing of beta portfolios.
Refinitiv’s report, Using Refinitiv News Analytics Over Monthly Horizons (which includes research notes from a previous white paper, Sentiment and Investor Behavior) describes how investors are able to gain potential benefits by using Refinitiv News Analytics.
Post-earnings announcement drift
Markets react strongly immediately after a large earnings surprise, and stock prices typically continue to drift in the same direction for the following weeks or months. This is called ‘post-earnings announcement drift’, or PEAD, and is usually caused by investors underreacting to the earnings surprise.
When an earnings surprise is contrary to market sentiment, investors are often slower or less likely to react. This causes a larger post-earnings announcement drift than when the earnings surprise is in line with market sentiment.
News sentiment index
Refinitiv News Analytics can be used to discover the extent of the drift, by constructing a news sentiment index. In this case, based on S&P 1500 companies appearing in Reuters news articles over the 30 days before the earnings surprise, as illustrated in the chart below.
The sentiment data demonstrates potential excess returns resulting from the post-earnings announcement drift and its expected timespan, opening the door for investors to profit from trading earnings surprises.
Pricing of beta portfolios
Another use of trailing sentiment data is in pricing risk.
Beta, which is used in the capital asset pricing model (CAPM), is a measure of the volatility, or systematic risk, of a security or portfolio, in comparison to the market as a whole. The relationship described by the model suggests higher-risk assets should generate higher rates of return.
Adding sentiment data, generated by Refinitiv News Analytics across the S&P 1500 universe, to the equation both proves and challenges the model.
Following periods of negative market sentiment, the risk-return relationship of the model plays out well, with higher beta portfolios generating excess returns, or alpha.
However, following periods of positive market sentiment, the relationship breaks down. Lower beta portfolios outperform higher beta portfolios that generate negative abnormal returns.
These outcomes suggest beta portfolios are priced appropriately following periods of negative market sentiment and mispriced following periods of positive market sentiment.
Armed with this information, investors can realize the profitability of incorporating news sentiment into their investment process.
These use cases of Refinitiv News Analytics confirm that market sentiment data can deliver real benefits, including a better understanding of investor behavior and financial profits.
What is Refinitiv News Analytics?
Refinitiv News Analytics is based on a natural language processing (NLP) engine and provides automated sentiment and linguistic analytics on financial news. Its quantitative news factors include news sentiment scores, relevance, novelty and volume:
- Sentiment scores suggest whether a news item conveys a positive, negative or neutral outlook on a particular asset.
- Relevance indicates the likelihood of a news item being relevant for a specific company or commodity.
- Novelty is the number of news items with similar content published over a trailing window.
- Volume counts how many times a particular asset has been mentioned in news over a trailing window.