A parsimonious quantile regression model to forecast day-ahead value-at-risk

This paper proposes a parsimonious quantile regression model for forecasting Value-at-Risk. The model uses only observable measures of daily, weekly, and monthly volatility as input and thus simplifies optimization substantially compared with other methods proposed in the literature. The framework also provides a new way of illustrating the volatility effects of a heterogeneous market. When subjected to formal coverage tests for out-of-sample VaR predictions, model performance is similar to more complicated models.