Analyst Reports and Stock Performance: Evidence from the Chinese Market
Rui Liu, Jiayou Liang, Haolong Chen, Yujia Hu
TL;DR
This paper analyzes whether sentiment in a large corpus of Chinese analyst reports predicts near-term stock performance. It trains a Chinese BERT-wwm model on $73836$ documents (training $$62735$$, test $11101$$) to produce sentiment scores $\hat{Pos}_{i,t}$ and $\hat{Neg}_{i,t}$, then regresses day-ahead stock metrics on these signals using $Range$, $Ret^{ex}$, and $\Delta volume$ with relevant market and firm controls. The key findings are that positive sentiment increases next-day excess returns and intraday volatility, while negative sentiment reduces excess returns but still raises volatility, with positive sentiment generally having a stronger impact. Robustness checks by industry and manual labeling largely corroborate the results, suggesting that sentiment extracted from professional Chinese analyst reports contains informative content about stock dynamics in the Chinese market. The work contributes to sentiment analysis in finance, demonstrates the viability of applying Chinese language models to financial texts, and offers implications for investors and researchers studying market reactions to expert commentary.
Abstract
This article applies natural language processing (NLP) to extract and quantify textual information to predict stock performance. Using an extensive dataset of Chinese analyst reports and employing a customized BERT deep learning model for Chinese text, this study categorizes the sentiment of the reports as positive, neutral, or negative. The findings underscore the predictive capacity of this sentiment indicator for stock volatility, excess returns, and trading volume. Specifically, analyst reports with strong positive sentiment will increase excess return and intraday volatility, and vice versa, reports with strong negative sentiment also increase volatility and trading volume, but decrease future excess return. The magnitude of this effect is greater for positive sentiment reports than for negative sentiment reports. This article contributes to the empirical literature on sentiment analysis and the response of the stock market to news in the Chinese stock market.