Brain Combined Sentiment

Product Summary 

The Brain Combined Sentiment dataset provides aggregated sentiment metrics for U.S. stocks, derived from multiple textual financial sources, including news articles, earnings calls, and filings. These metrics are calculated using different Brain datasets and combined into a single file for the convenience of investors interested in stock sentiment analysis.

Sentiment Metrics

The provided sentiments metrics are the following:

1. Sentiment and volume from news calculated over the past 30 days.

2. Sentiment from the most recent earnings call of the Management Presentation section.

3. Sentiment from the most recent 10-K filing calculated on the whole document.

4. A combined sentiment score calculated as the average sentiment across all sources.

Each input source and its associated sentiment operates on a distinct time scale: sentiment from news is aggregated over the past month, sentiment from earnings calls is updated quarterly, and sentiment from 10-K filings reflects an annual time scale. The sentiment of each source is normalized in a point-in-time, walk-forward manner to enable the comparison and combination of the various sentiment scores from various sources.

Temporal Context

For each source, the dataset also includes the temporal context of the associated sentiment data, specifically: the number of days used to aggregate sentiment from news, the number of days since the most recent earnings call, and the number of days since the most recent filing report. The temporal information can be used by the user for further weighting each sentiment source according to their specific requirements.

Coverage

The dataset includes all US stocks.