Quantify the impact of media coverage

Incorporate the effect of public information into your investment decisions from the systematic analysis of 40,000 curated news and social media sources.

2004-01-01 2023-08-01

Every day, thousands of news stories are published on publicly listed companies. Each one may move their stock price, but only a systematic approach can capture the sentiment momentum. RavenPack Company News Factors provide a simple, intuitive, and robust approach to incorporate this signal into strategies and dashboards.

Statistically and Economically Significant

News Sentiment and Media Attention Insights

Company News Factors can help quantitative and discretionary investors:

See it in action

Signal

Event Group (Event sentiment only)

Sectors

Top companies

Bottom companies

Backed by Research

While RavenPack Company Factors can be leveraged as features in a broad variety of investment strategies, the signal components have been used to harness alpha in specific contexts as demonstrated by research. Below are examples of such usage:

In Multi-Factor Strategies

Like any other factors, RavenPack Company News Factors can serve as an input to a trading model. The Factors prove particularly effective at ranking stocks, and research has established the relevance of sentiment even as a stand-alone factor for trading.

Time Value Buys

When trading the Value factor, you may identify a bargain stock whose price is eroded by negative news. RavenPack Company Factors can help you use sentiment and media attention to determine whether the market has fully priced the effect of negative news, or if you can expect further price erosion that improves your opening position.

Time Momentum Sells

If you trade on Momentum, you can optimize your profile by flattening your position when the stock is reaching the end of its run. Studies have identified that sentiment can enhance this decision-making process.

Apply Overlays

Existing factors can benefit from the addition of a sentiment overlay as a condition or a tilt. For instance you can overweight or underweight the trades of a model depending on the RavenPack Company News Factor signal.

Qualify Shorts

Companies with significant short interest tend to underperform, but their tactical performance profile depends on the rationale for investors’ negative outlook on the stock. RavenPack Company Factors provide a quantitative view of media attention and sentiment that can better inform your analysis —either by attributing the short positions to investors’ interpretation of public news volume and direction, or as the result of their holding non-public information which can drive longer-term pressure on the stock.

Build Thematic Signals

By tracking sentiment tied to themes like earnings, analyst ratings or product and services, you gain more insights into events driving the markets and make more informed trading decisions. Not all types of events carry the same market reactions, nor do they interact in the same way with other factors. The ability to separate sentiment into themes brings more flexibility when incorporating these factors into your strategies.

'
UNDER THE HOOD

Computing the Factors

Company News Factors are the outcome of a multi-step process to turn unstructured textual content into actionable features:

Content Collection

RavenPack analyzes over 300 million news and social content per month from 40,000 sources to identify information relevant to publicly listed companies.

Information Extraction

Using a powerful taxonomy, RavenPack then sorts events using categories like acquisitions & mergers, analyst ratings, assets, credit, credit ratings, dividends, earnings, equity actions, insider trading, labor issues, legal, marketing, partnerships, price targets, products services, regulatory, revenues, sanctions, stock picks, and stock price.

Analytics Enrichment

Both documents and events are then enriched with metadata that helps determine the relevance and novelty of the information, and can further describe its context and meaning, including sentiment scores.

Factor Construction

Two types of factors are then constructed for each company:

  • document-level factors, which aggregate analytics on any document that mentions the company with sufficient relevance and novelty, as used to report media attention, and
  • event-level factors, which specifically identify and aggregate detected events of sufficient relevance and novelty for the company.
Company Factors Facts
Purpose
  • Harness sentiment alpha
  • Mitigate portfolio risk
  • Leverage across asset classes
Frequency6 times daily
HistorySince 2001
Universe100,000+ Listed Companies
Content
  • Company-related events factors inform on market-moving news
    • For each event group, the volume of events that report that group, and the average sentiment score associated to that group for that company are provided.
    • Additional fields describe event novelty, relevance, score, and other metadata.
  • Company-related documents factors inform on media attention
    • For each company, the volume of documents where it is being mentioned, and the average sentiment score associated with them are provided.
    • Additional fields describe document novelty, relevance, score, and other meta data.
Delivery RavenPack, Snowflake

Access the Company News Factors


By submitting this form, you agree to RavenPack's terms of service and privacy policy.