Volatility Analysis Using High-Frequency Financial Data

Stock market, whose total size exceeds 10 billion dollars, have boomed in the past few decades, becoming a crucial indicator of global economy. It is universally acknowledged that high-frequency data, stock price for instance, fluctuates dramatically during market crash as well as other financial events, creating numerous volatility clusters and jumps, which makes volatility analysis arduous and burdensome. Based on previous academic researches concerning Time Serie Momentum and Asset Pricing, I select several typical days with extreme financial events and conduct some empirical works such as analyzing RRV distribution of those days and calculating correlation coefficients, concluding four characteristics of the data including irrelevance, fat-tail and asymmetry, leverage effect and volatility clustering, and categorizing them to better unfold its overall distribution. My statistical works provide stock investors with an exhaustive and clear overall understanding concerning stock price volatility, helping them make better investment decisions and eventually receive better return during their stock investment.