Title: Bagged value-at-risk forecast combination
Authors: Ekaterina Kazak - University of Manchetser (United Kingdom) [presenting]
Roxana Halbleib - University of Freiburg (Germany)
Winfried Pohlmeier - University of Konstanz (Germany)
Abstract: Recent developments in financial econometrics literature on joint scoring functions for Value-at-Risk and Expected Shortfall allowed for consistent implementation of statistical tests based on the Model Confidence Set (MCS). MCS is shown to be a great tool for model comparison, both in-sample and out-of-sample. Another branch of literature focused on the superior performance of convex forecast combinations, which often outperform stand-alone forecasting models. Both results are combined, and a novel approach is proposed to a forecast combination of Value-at-Risk and Expected Shortfall based on the MCS. We exploit the statistical properties of bootstrap aggregation (bagging) and combine competing models based on the bootstrapped probability of the model being in the Confidence Set. The resulting forecast combination allows for a flexible and smooth switch between the underlying models and outperforms the corresponding stand-alone forecasts.