Dynamic factor Value-at-Risk for large heteroskedastic portfolios

- We propose an efficient VaR methodology based on historical simulation and dynamic factor models.
- The methodology is suitable for large portfolios with time-varying volatilities and correlations.
- We test the methodology on stock portfolios with different distributional characteristics.
- DFM-VaR compares well to historical simulation and univariate filtered historical simulation VaRs.

We propose a methodology that can efficiently measure the Value-at-Risk (VaR) of large portfolios with time-varying volatility and correlations by bringing together the established historical simulation framework and recent contributions to the dynamic factor models literature. We find that the proposed methodology performs well relative to widely used VaR methodologies, and is a significant improvement from a computational point of view.