This article explores realized volatility—volatility measured from intraday returns—and the Heterogeneous Autoregressive (HAR) model. It explains why HAR-RV outperforms GARCH for short-term forecasts, the role of jumps and leverage effects, and how traders use 5-minute data to predict tomorrow's risk.

This article explains the GARCH framework, which models volatility as a function of past squared returns and past volatilities. It covers volatility clustering, mean reversion, and the key differences between ARCH, GARCH, and EGARCH. Applications include value-at-risk (VaR), option pricing, and dynamic hedging.