A Study on Asymmetric Leverage Effects Based on GARCH: The Case of the Artificial Intelligence Sector

DOI: https://doi.org/10.62517/jse.202611311

Author(s)

Qian Zhou

Affiliation(s)

East China University of Science and Technology, Shanghai, China

Abstract

Financial asset returns often exhibit asymmetric reactions to positive and negative news, a phenomenon known as the leverage effect, which is particularly pronounced in high-tech growth sectors. This study examines the China Securities Artificial Intelligence Sector Index (931071) using daily return data from October 2020 to October 2025. GARCH, TGARCH, and EGARCH models are constructed to systematically investigate the volatility characteristics and asymmetry of returns in the AI sector. Findings reveal: The standard GARCH(1,1) model demonstrates optimal goodness-of-fit and diagnostic performance, indicating volatility clustering as the primary driver of sector fluctuations. However, TGARCH and EGARCH models exhibit divergent estimates for asymmetry parameters-the former showing a non-significant negative leverage effect, while the latter exhibits a significant positive effect, suggesting positive news may trigger stronger volatility in this sector. This anomalous effect may be closely linked to the sector's high valuations, sentiment-driven dynamics, and policy sensitivity. The findings provide theoretical and empirical support for understanding the volatility mechanisms of high-tech assets and optimizing risk management strategies.

Keywords

GARCH Model Family; Asymmetric Leverage Effect; TGARCH Model; EGARCH Model

References

[1] Challa M L, Malepati V, Kolusu S N R. S&P BSE Sensex and S&P BSE IT return forecasting using ARIMa[J]. Financial Innovation, 2020,6(1):1-19. [2] Devi B U, Sundar D, Alli P. An effective time series analysis for stock trend prediction using ARIMa model for Nifty Midcap-50[J]. International Journal of Data Mining & Knowledge Management Process, 2013, 3(1):65-78. [3] Green J, Hand J R M, Zhang X F. The characteristics that provide independent inforMation about average U.S. monthly stock returns[J]. The Review of Financial Studies, 2017, 30(12):4389-4436. [4] LinnainMaa J T, Roberts M R. The history of the cross-section of stock returns[J]. The Review of Financial Studies, 2018, 31(7):2606-2649. [5] Gupta R, Pierdzioch C, Vivian A J, et al. The predictive value of inequality measures for stock returns: An analysis of long-span UK data using quantile random forests[J]. Finance Research Letters, 2019, 29:315-322. [6] Huang W, Nakamori Y, Wang S Y. Forecasting stock Market movement direction with support vector Machine[J]. Computers & operations research, 2005, 32(10):2513-2522. [7] Bollerslev T. Generalized autoregressive conditional heteroskedasticity [J]. Journal of econometrics,1986,31(3):307-327. [8] Crouhy M, Rockinger M. Volatility clustering, asymmetry and hysteresis in stock returns:International evidence[J]. Financial engineering and the Japanese Markets,1997,4(1):1-35. [9] Zhang Shiying, Li Handong, Fan Zhi. The Persistence of Financial Risk and Its Hedging Strategies [J]. Systems Engineering - Theory & Practice, 2002(5): 31-36. [10] Li Handong, Zhang Shiying. Research on Persistence and Co-persistence in Stochastic Volatility Models [J]. Journal of Systems Engineering, 2002(4): 289-295+302. [11] Xu Lixia. Research on the Application of Volatility Models in China's Stock Market [J]. Statistics and Decision, 2010(12): 168-170. [12] Ren Chengke, Jiang Xiaogan. Current Status and Development Analysis of GARCH Modeling in Financial Time Series [J]. Times Finance, 2012(9): 153-154.