报告题目： Functional Quantile Autoregression

报告人：董朝华教授

时间：12月11日（周一）上午10:00-11:00                                 

地点：创新港涵英楼经济金融研究院8003报告厅

报告人简介：

董朝华，现为中南财经政法大学统计与数学学院教授，博士生导师，澳大利亚阿德莱德大学经济学博士、莫纳什大学博士后。研究兴趣包括高维计量理论、非参数与半参数方法、非平稳时间序列和面板数据模型、金融时间序列与微观计量应用。研究成果发表于Journal of Econometrics, Econometric Theory, Journal of Business & Economic Statistics, JRSSB, Annals of Statistics等计量经济学与统计学顶级期刊，主持国家自科项目两项。

摘要：

In this paper, we propose a new class of time series models, the functional quantile autoregression (FQAR) models, in which the conditional distribution of the observation at the current time point is affected by its past distributional information, and expressed as a functional of the past conditional quantile functions. The models can capture systematic influences of the past distributional information on the current conditional distribution. The FQAR models are similar to functional autoregression models. However, different from conventional functional time series models which are based on functionally observed data, the proposed model and method study functional dynamics in traditional time series data. In fact, the FQAR models share some similar features to the GARCH model, while, instead of only focusing on the dynamic relationship of the second moments (conditional variance), we look at the relationship between the current conditional distribution to the past conditional distribution in a functional autoregressive form. Stationary conditions of the model are provided. We propose a sieve estimator for the model. Under stationarity conditions, the FQAR model converges and facilitates an estimation that does not depend on the initial quantile function. Identification is investigated. The asymptotic properties of the estimators are derived. A Monte Carlo experiment is conducted to investigate the finite sample performance of the proposed estimator and an empirical application to stock return time series highlights the proposed method.