Weibo Emotions and Chinese Stock Returns Across Time and Frequency

Summary

The paper examines how five emotions expressed on Sina Weibo—anger, disgust, fear, joy, and sadness—relate to Chinese stock returns. It uses machine learning to classify posts and wavelet analysis to study how the relationship changes over time and across frequencies, rather than treating sentiment and returns as having one fixed correlation.

The reported relationship is positive during periods of notable return fluctuations at medium to higher frequencies, especially for horizons under ten trading days from October 2014 onward. Sadness appears more closely associated with returns than the other emotions. In most periods with significant links, returns precede changes in Weibo emotion; the study also reports that market movements affect optimism and agreement, with agreement associated with returns moving together over roughly 40 trading days. These are findings from the analyzed sample, not evidence of a durable trading signal or causal effect that will generalize to other periods or markets.

Key ideas

• Machine learning is used to classify Weibo posts into five emotion categories.
• Wavelet analysis reveals that the sentiment-return relationship varies over time and frequency.
• The study reports positive links between emotions and returns during some periods of short-horizon fluctuations.
• Stock returns generally precede changes in social media emotions in the periods with significant connections.
• The reported associations do not establish a reliable or generalizable trading strategy.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.