Investor Sentiment, Cross-Market Contagion, and Asset Pricing in China: A Sentiment-Extended Fama–French Framework

Investor sentiment can transmit non-fundamental price pressure across markets, yet standard asset-pricing models do not explicitly capture the interaction between broad market sentiment, high-frequency attention, and narrative shocks. This paper develops a China-adapted Sentiment-Extended Fama–French Six-Factor model (CASE-FF6) for A-share equities. The framework combines a lagged Baker–Wurgler-style market sentiment component, NLP-derived financial-news tone, and abnormal Baidu search intensity into a hybrid sentiment state variable. After orthogonalization against conventional factors and macroeconomic controls, the residual signal is converted into a tradable sentiment-mimicking factor and added to a China-adapted Fama–French five-factor baseline. The proposed model is evaluated through nested comparisons with FF3, sentiment-augmented FF4, and FF5 specifications, using time-series pricing errors, the Gibbons–Ross–Shanken test, Fama–MacBeth cross-sectional regressions, and out-of-sample performance. A complementary event framework uses high-salience Nvidia-related announcements in 2026 to examine herding, attention-driven buying, and liquidity amplification in Chinese concept stocks. The study therefore distinguishes sentiment measurement from factor pricing and provides a falsifiable design for testing whether sentiment represents a distinct priced return component or a temporary conditioning mechanism in China's equity market.