Abstract

The heterogeneous autoregressive (HAR) model is extended by modeling the joint distribution of the four partial-volatility terms therein involved. Namely, today's, yesterday's, last week's and last month's volatility components. The joint distribution relies on a (C-) Vine copula construction, allowing to conveniently extract volatility forecasts based on the conditional expectation of today's volatility given its past terms. The proposed empirical application of the novel CV-HAR model involves more than seven years of high-frequency transaction prices for ten stocks and evaluates the in-sample, out-of-sample and one-step-ahead forecast performance of the CV-HAR model for daily realized-kernel measures of volatility. The proposed new forecasting model is shown to outperform the HAR benchmark under different models for marginal distributions, copula construction methods, and forecasting settings.