Incorporating sentiment into investment strategies
Discovering the value of adding sentiment to multi-factor equity strategies
Sentiment in news and social media is increasingly seen as an invaluable addition to more traditional investment signals.
News impacts security prices through sentiment
Media content and its sentiment is an alternative data source that is currently gaining popularity in the investment community. This data can be used to seek additional return as well as to limit risk of investment portfolios.
Generally, when news regarding stock becomes more positive, increasing its sentiment, more investors consider buying the stock, resulting in a positive impulse to its price. Therefore, the change of sentiment is the price driver, rather than its level.
In this paper, Probability & Partners use Refinitiv News Analytics to investigate change in sentiment, combining sentiment with other quantitative news characteristics to capture the influence of news media on stock prices.
The main objective is to answer the question of whether there is added value in incorporating sentiment in existing investment models.
Access the full report to discover:
- Ways to Incorporate sentiment into factor models - Findings show that incorporating sentiment into traditional factor strategies produces additional gains for returns without increasing risk.
- How to use sentiment as a risk overlay - By avoiding stocks whose sentiment is declining, returns of investment strategies are enhanced.
- Various ways to use sentiment - Combining the use of sentiment in several ways achieves the greatest portfolio performance.
35,000equities covered by Refinitiv News Analytics
90+metadata fields covered by the product
2003start of our historical news and analytics scores
Read on - Complete this form to open the full report
These findings demonstrate the persistent impact of high-quality news on equity markets. It starts with Reuters News, providing timely and trusted news and ends with the Refinitiv News Analytics engine providing high-quality sentiment and entity relevance scores.