Heterogeneous autoregressive model with structural break using nearest neighbor truncation volatility estimators for DAX

Abstract

High frequency financial data modelling has become one of the important research areas in the field of financial econometrics. However, the possible structural break in volatile financial time series often trigger inconsistency issue in volatility estimation. In this study, we propose a structural break heavy-tailed heterogeneous autoregressive (HAR) volatility econometric model with the enhancement of jump-robust estimators. The breakpoints in the volatility are captured by dummy variables after the detection by Bai–Perron sequential multi breakpoints procedure. In order to further deal with possible abrupt jump in the volatility, the jump-robust volatility estimators are composed by using the nearest neighbor truncation approach, namely the minimum and median realized volatility. Under the structural break improvements in both the models and volatility estimators, the empirical findings show that the modified HAR model provides the best performing in-sample and out-of-sample forecast evaluations as compared with the standard HAR models. Accurate volatility forecasts have direct influential to the application of risk management and investment portfolio analysis.

A Bayesian analysis of complete multiple breaks in a panel autoregressive (CMB-PAR(1)) time series model

“…The consequences of the structural break under Bayesian approach is studied by several researchers, see Albert and Chib (1993), Bai (2010), Kumar et al (2012), Eo (2012) and Maheu and Song (2018). Further on, Chin et al (2016) combined both robust-jump volatility estimator and a structural break heterogeneous autoregressive (HAR) model to battle the structural break in stock market volatility modelling and added the empirical literature of high-frequency volatility analysis by using modified HAR models and robust-jump volatility estimators. Yamamoto (2016) considered a simple modification in EM confidence set proposed by Elliott and Muller (2007) in a linear regression model having a single structural break and achieved a shorter confidence set than the EM method.…”