Investor sentiment, registration reform, and post-IPO performance in China’s chinext market

This study examines how investor sentiment and China’s registration- based IPO reform jointly shape IPO pricing and post-listing performance in the ChiNext market. Using a sample of 704 IPOs from 2016 to 2022, this study constructs a composite sentiment index by combining market- based indicators with textual sentiment extracted from online investor comments using a machine learning approach. The results show that stronger pre-IPO sentiment is associated with higher initial returns, indicating sentiment-driven overpricing prior to listing. In contrast, the registration-based reform is linked to lower initial underpricing, suggesting improved price discovery under a more market-oriented issuance system. Over longer horizons, higher sentiment predicts significantly negative abnormal returns, consistent with the correction of initial overvaluation. In addition, elevated sentiment is associated with higher post-listing volatility, particularly in the early aftermarket period. Overall, the findings suggest that investor sentiment and registration-based reform jointly influence IPO outcomes, with institutional reform mitigating mispricing and enhancing market efficiency. This study contributes to the literature by providing evidence on the interaction between behavioral factors and institutional change, and by demonstrating the value of machine learning-based sentiment measures in the analysis of IPO performance.