Impacts Of Macroeconomic Variable On Stock Market Volatility

As one of the cores of modern financial theory, volatility becomes a measure of investment risk. It’s a hot topic to accurately forecast volatility in the study of modern finance. Low-frequency data is often used to forecast volatility in existing model, but inevitably loss a lot of information. With the development of computer technology and the exploration of the high-frequency data model, it is possible to calculate the "realized estimators" such as the realized volatility and realized bi-power variation by using high-frequency data modeling. By analyzing the "realized estimators", it can overcome the shortcomings resulting from low-frequency modeling. Based on the separation of realized volatility, I will further distinguish the continuous and jump components, and select the HAR-CV-CJ model.Considering that there is close correlation between the macroeconomic and stock market volatility, I intend to choose these macroeconomic variables such as industrial added value、the total value of import and export commodities, Consumer Price Index, interest rate and exchange rate to join the HAR-CV-CJ model. Before that, I will analysis the principal components of the five macroeconomic variables and extract the two main components:investment positive components and negative components.Compare the original model with the model after introducing the macroeconomic variables, I find that the new model is better at fitting ability and has better data modeling results after being taken logarithm. Macroeconomic variables have weaker impact on the realized volatility than continuous volatility. And, the investment positive components have more significant impact than the investment negative components.