Dynamic conditional correlations, forecasting volatility and time-varying effects between macroeconomic variables and stock market volatility in China

The Chinese stock market plays a pivotal role in the healthy development of the national economy and in the integration of the global economy. The effective prevention of risks and the assurance of a safe and stable operation of the stock market are particularly crucial, necessitating an accurate depiction of the stock market’s volatility characteristics. Gaining a deep understanding of dynamic conditional correlations, forecasting volatility, and the time-varying relationship between macroeconomic variables and stock market volatility is crucial for ensuring the healthy and stable development of the Chinese stock market. The purpose of this research involves examining the relationship between macroeconomic variables and stock market volatility: this involves exploring the impacts of macroeconomic variables on the volatility of the Chinese stock market using the Time-varying Parameter Structural Vector Autoregression (TVP-VAR) model with stochastic volatility and empirically examining the dynamic time-varying linkage between macroeconomic variables and stock market volatility; utilizing low-frequency (macroeconomic) information aims to enhance the prediction accuracy of high-frequency stock market volatility and to explore the decomposition of stock market volatility along with its macroeconomic roots; employing low-frequency (macroeconomic) information and high-frequency stock market volatility, this research seeks to investigate the dynamic connections between macroeconomic variables and stock markets. This research focuses on the characteristics of stock market volatility and employs econometric methods to delve into the relationship between stock market volatility and macroeconomic variables. It can serve as a foundational reference for the government in formulating macroeconomic policies, help to deepen the understanding of macroeconomic variables, and illuminate the dynamic relationship between the stock market and the academic community. Additionally, it offers guidance for investors to adjust long-term investment strategies, optimize asset portfolios, and enhance investment returns. This study provides empirical evidence that the TVP-VAR model outperforms the VAR model in evaluating the dynamic time-varying linkage between macroeconomic variables and stock market volatility, extending the literature on the impact of macroeconomic variables on stock market volatility. The GARCH-MIDAS model outperforms the standard GARCH (1,1) model in volatility forecasting performance when considering the influence of macroeconomic variables, further extending the literature on estimation and forecasting performance, particularly in comparing level effects and volatility effects. Additionally, this study introduces an innovative methodology, Dynamic Conditional Correlation-Generalized Autoregressive Conditional Heteroscedasticity-Mixed Data Sampling (DCC-GARCH-MIDAS). The DCC-GARCH-MIDAS model outperforms the DCC-GARCH model in volatility forecasting performance when considering the influence of macroeconomic variables. This study extends the literature on the dynamic correlation between macroeconomic variables and China’s three major stock markets and represents the first attempt to evaluate the performance of the DCC-GARCH-MIDAS model in volatility forecasting, using realized volatility (RV) as a proxy for actual volatility. Benefitting from accurate estimation and prediction, this research offers a comprehensive interpretation of the volatility effects.

Text 125409 - Full Text.pdf Download (6MB) Text SPE 2024 16 - Declaration Form.pdf Restricted to Repository staff only Download (443kB) Text SPE 2024 16 - Full Text.pdf Restricted to Repository staff only Download (6MB) Official URL or Download Paper: http://ethesis.upm.edu.my/id/eprint/18881

Actions (login required)

View Item