Volume 4,Issue 6
投资者舆情情绪极性对股票收益的中介效应研究—— 基于多模型融合的创新视角
本研究突破传统单一模型分析框架，创新性地融合Baron-Kenny 中介效应模型、Sobel 检验、Bootstrap 方法、GARCH 族模型、面板向量自回归（PVAR）以及门限回归等多种前沿计量方法，构建了一个多维度、多层次的分析体系，深入探讨投资者舆情情绪极性对股票收益的影响机制与传导路径。研究发现：（1）波动率在舆情情绪与股票收益之间起到了显著的完全中介作用，中介效应占比高达1266.05%，揭示了舆情影响市场的核心传导渠道；（2）通过GARCH、EGARCH 和TGARCH 模型的对比分析，发现舆情情绪对条件波动率存在非对称影响，负面舆情的冲击效应更为显著；（3）面板向量自回归分析表明，舆情情绪、收益率和波动率之间存在复杂的动态交互关系，形成了一个完整的反馈回路；（4）门限回归分析发现，当舆情情绪低于门槛值时，中介效应更为显著。
[1]Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction
in social psychological research: Conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51(6), 1173-1182.
[2]Preacher K J, Leonardelli G J. Calculation for the Sobel test[J]. Retrieved January,
2001, 20: 2009.
[3]Rajh-Weber H, Huber S E, Arendasy M. Using heteroskedasticity-consistent
standard errors and the bootstrap for linear regression analysis available in SPSS: A
tutorial[J]. Advances in Methods and Practices in Psychological Science, 2026, 9(1):
25152459251408046.
[4]Bollerslev T, Engle R F, Nelson D B. ARCH models[J]. Handbook of econometrics,
1994, 4: 2959-3038.
[5]Bhat P A A R, Shakila B, Pinto P, et al. Comparing the performance of GARCH
family models in capturing stock market volatility in India[J]. Journal of Management,
2034, 11(3): 11-20.
[6]Holtz-Eakin D, Newey W, Rosen H S. Estimating vector autoregressions with
panel data[J]. Econometrica: Journal of the econometric society, 1988: 1371-1395.
[7]Wang Y. Asymptotic nonequivalence of GARCH models and diffusions[J]. The Annals
of Statistics, 2002, 30(3): 754-783.
[8]Apostolakis G, Papadopoulos A P. Financial stability, monetary stability and
growth: a PVAR analysis[J]. Open Economies Review, 2019, 30(1): 157-178.
[9]Amaro M, Molontay R. From Beijing to Paris: Media representation and sentiment
dynamics of the Olympics and Paralympics across traditional and social platforms
(2008–2024)[J]. International Journal of Sports Science & Coaching, 2026:
17479541261435444.