Change point detection for the intraday volatility using functional ARCH and conditional Copula

In this research, we are concerned with intraday volatilities computed by functional ARCH(1) (fARCH(1), for short) model for high-frequency financial time series. A conditional-Copula multiple change point detection (CPD) for intraday volatilities is proposed using fARCH(1), bivariate Gaussian Copula and t-Copula conditional distributions. We employ current available multivariate CPD models which include energy test based control chart (ETCC) and nonparametric multivariate change point model (NPMVCP) to implement the proposed CPD method for the intraday volatilities. A simulation study is conducted to demonstrate that the functional ARCH based conditional-Copula CPD for the intraday volatilities can be a useful econometrics method to detect abnormal intraday volatilities in the financial market. We analyze intraday volatilities of the Korea composite stock price index (KOSPI) and the Hyundai-Motor (HDM) company stock data with one minute high-frequency to illustrate our proposed CPD method.