A0172
Title: Bregman model averaging for forecast combination
Authors: Chu-An Liu - Academia Sinica (Taiwan) [presenting]
Yi-Ting Chen - National Taiwan University (Taiwan)
Jiun-Hua Su - Academia Sinica (Taiwan)
Abstract: A unified model averaging (MA) approach is provided, and its asymptotic optimality is established for a wide class of forecasting targets. The asymptotic optimality is achieved by minimizing an asymptotic risk based on the expected Bregman divergence of a combined-forecast sequence from a forecasting-target sequence under the local(-to-zero) asymptotics. This approach is flexibly applicable to generate MA methods in different forecasting contexts, including, but not limited to, univariate or multivariate mean forecasts, volatility forecasts, probabilistic forecasts and density forecasts. We also conduct Monte Carlo simulations (empirical applications) to show that compared to related existing methods, the MA methods generated by this approach perform reasonably well in finite samples (real data).
