Answers are grounded in this paper and cite their sources.
Purpose This study investigates the non-linear and asymmetric relationship between investor sentiment and equity returns in China, examining how sentiment influences market-wide and sectoral returns under different market conditions. It seeks to establish whether sentiment–return relationships vary across bullish, bearish and neutral regimes and across industries with differing characteristics. By focusing on China’s distinctive institutional environment, the study contributes to understanding investor sentiment as a state-dependent behavioural predictor of equity returns rather than necessarily a priced systematic risk factor. Design/methodology/approach Using monthly data from January 2014 to December 2023, the study examines the CSI 300 Index and ten sectoral indices. Investor sentiment indices are constructed following Baker and Wurgler (2006), using principal component analysis (PCA). Quantile-on-quantile (QQ) regression captures non-linear dependencies across the joint distributions of sentiment and equity returns, incorporating macroeconomic control variables. Ordinary least squares regression provides a baseline comparison. Robustness is assessed using an alternative sentiment index constructed via partial least squares (PLS). Findings Investor sentiment exhibits a non-linear, asymmetric and state-dependent relationship with Chinese equity returns. Positive associations predominate during normal and optimistic market conditions, whereas negative relationships emerge when extreme pessimism coincides with severe market downturns. The magnitude and direction of these effects differ considerably across sectors, reflecting heterogeneous sensitivities to market conditions. Sentiment effects are particularly pronounced at extreme quantiles, indicating their economic relevance beyond conventional statistical significance. Results using the PLS-based sentiment index broadly corroborate the PCA-based findings, supporting the robustness of the principal sentiment–return patterns. Originality This study extends behavioural finance research by applying QQ regression to examine investor sentiment and equity returns simultaneously across market-wide and ten sectoral distributions in China. Unlike conventional approaches focusing on average effects, it identifies heterogeneous sentiment–return relationships across market regimes and industries. The research contributes evidence from a major emerging equity market characterised by distinctive institutional arrangements and substantial retail investor participation. By distinguishing conditional return predictability from priced systematic risk, it also offers a more theoretically precise interpretation of sentiment’s role in equity markets.
No takes yet. Share an insight, caveat, or question.
Bouteska et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: