Daily Twitter sentiment scores

What is Context Analytics?

Context Analytics is a financial technology company that turns social media activity into sentiment signals for tradable securities. It runs a patented natural language processing engine over tweets referencing equities, ETFs, commodities, currencies, and other securities, scoring each post for sentiment and then aggregating those scores into daily metrics researchers can compare against a security's own baseline.

‍

What academic researchers should know about Context Analytics social sentiment data

On Dewey, Context Analytics publishes Social Sentiment Metrics for Equities, a daily feed of fifteen sentiment and volume-based metrics per ticker. Documented coverage spans about 4,500 U.S. equities, 450 TSX listed Canadian equities, 1,250 LSE FTSE listed U.K. stocks, 2,500 ETFs, 113 commodities, 49 currency pairs, about 750 cryptocurrencies, and 150 top private companies. Each metric is built from a rolling 24 hour window of tweets from accounts Context Analytics has rated as credible, then compared against a 20 day moving average and volatility baseline to produce scores like S-Score and S-Buzz that flag unusual sentiment or volume for a given ticker.

‍

Why academic researchers choose Context Analytics on Dewey

Most social media sentiment research starts by scraping and scoring raw text, which is a substantial data engineering project before any actual analysis can begin. Context Analytics does that scoring work up front with a patented, multi pass natural language processing pipeline, so researchers get a ticker level daily panel of sentiment and volume metrics they can join directly to price and volume data. Because the metrics are already benchmarked against each security's own rolling baseline, researchers can identify unusual sentiment or attention without having to build that normalization themselves. It pairs naturally with other Dewey partners for asset pricing and fixed income research: Exchange Data International for global end of day equity pricing, and 7 Chord for issuer level credit curves.

‍

Context Analytics academic research ideas and use cases

• Sentiment and short term returns. Daily S-Score and S-Buzz readings can be tested against next day or next few day returns and volatility for the same tickers, following the logic of the broader academic literature linking unusual social media sentiment to short horizon price moves.
• Sentiment versus attention. Because the feed separates volume based metrics like S-Volume and S-Buzz from sentiment based metrics like S-Score, researchers can disentangle whether a move is driven by more people talking about a stock or by the tone of what they are saying.
• Cross-asset sentiment spillovers. Coverage across equities, ETFs, commodities, currency pairs, and cryptocurrencies supports research on whether sentiment in one asset class leads or lags sentiment in another.
• Behavioral finance and herding. The tweet source diversity metric, S-Dispersion, offers a way to test whether unusual sentiment reflects a broad consensus forming across many accounts or a smaller, more concentrated group driving the conversation.

‍