Title: Model averaging estimation for conditional heteroscedasticity model family
Authors: Qingfeng Liu - Otaru University of Commerce (Japan) [presenting]
Qingsong Yao - Renmin University of China (China)
Guoqing Zhao - Renmin University of China (China)
Abstract: The model averaging estimation for the conditional heteroscedasticity model family is considered. Given a set of candidate models with different functional forms, we propose a model averaging estimator for the conditional volatility and construct the corresponding weight choosing criterion. According to our results, the weight that minimizes the weight choosing criterion asymptotically minimizes the true KL divergence, as well as the Itakura-Saito distance.