A generalized heterogeneous autoregressive model using market information

This paper introduces a novel class of volatility forecasting models that incorporate market realized (co)variances and semi(co)variances within the framework heterogeneous autoregressive (HAR) model. Our empirical analysis shows statistically economically significant gains. For our most parsimonious market-HAR specification, stock is improved by 9.80% points. Using mixed sampling frequency variant with low (high) for (market) improves further 6.90% also develops noise-robust estimators to facilitate use at high frequencies.