Empirical Analysis of Stocks Based on ARIMA-GARCH-M Model

As the core of the modern Chinese economy, finance is an important tool for resource allocation and macro-control as well as an important force in promoting economic and social development. Since the 18th National Congress of the CPC, General Secretary Xi Jinping has attached great importance to the position and role of finance in economic development and social life, not only giving important speeches on issues such as financial reform, financial development and financial security but also providing important instructions for avoiding financial risks. As far as the stock market in the financial field is concerned, the trend of share price directly affects the economic interests of investors as well as indicates the country's economic policies. Due to the non-stationary time series of share price, the stock indexes are volatile. Under such circumstances, investors are exposed to investment risks and the stock market may experience abnormal fluctuations at any time. Therefore, accurate forecasting has become a challenging problem. In this article, the author intends to combine the ARIMA Model with the GARCH-in-MEAN Model based on previous studies to form the ARIMA-GARCH-M Model and then introduce it into the stock market to forecast the closing prices of SSE Index and HSI. Besides, the author not only tests and analyzes the volatility of SSE Index and HSI but also tests the feasibility of ARIMA-GARCH-M Model in stock index forecasting. Finally, with Granger causality test, the effect of mutual influence between Hong Kong stock market and Shanghai market is tested, and the experimental results are analyzed and summarized in the context of current macroeconomic environment.