A0784
Title: Return and volatility forecasting in mixed panels
Authors: Cindy Shin Huei Wang - HSBC Business School, Peking University (China) [presenting]
Abstract: A simple pooling prediction is proposed for a mixed panel model including stationary $I(0)$ and $I(d)$ processes via a pool autoregressive approximation (PAR) framework. This PAR-forecasting approach does not require prior information on the exact order and fractional parameter of each series of the mixed panel. It is also shown this approach remains valid when there exist common factors in a mixed panel. Insights from the theoretical analyses are confirmed by a set of Monte Carlo experiments, through which it is demonstrated that the approach outperforms existing forecasting methods. In particular, several controversial arguments in forecasting literature are also justified. Moreover, an empirical application to return and volatility forecasting illustrates the usefulness and feasibility of the forecasting procedure in portfolio settings.
