系列讲座

首页 > 系列讲座 > 正文

系列讲座

Estimating and Forecasting Volatility using Leverage Effect & Nonlinear regression with nonstationarity and heteroscedasticity

时间:2019-11-21

Statistics Seminar (2019-25)

Title: Estimating and Forecasting Volatility using Leverage Effect

Speaker: Christina Dan Wang, NYU Shanghai

Time: Thursday,November 28, 14:00-15:00

Place: Room 217, Guanghua Building 2

Abstract:

This research provides a theoretical foundation for our previous empirical finding that leverage effect has a role in estimating and forecasting volatility. This empirics is also related to earlier econometric studies of news impact curves (Engle and Ng, Chen and Ghysels). Our new theoretical development is based on the concept of projection on stable subspaces of semimartingales. We show that this projection provides a framework for forecasting (across time periods) that is internally consistent with the semi-martingale model which is used for the intra-day high frequency asymptotics. The paper shows that the approach provides improved estimation and forecasting both theoretically, in simulation, and in data.

Introduction:

Christina Dan Wang is Assistant Professor of Finance, NYU Shanghai; Global Network Assistant Professor, NYU. Prior to joining NYU Shanghai, Wang was an Assistant Professor in the Department of Statistics at Columbia University and a Postdoctoral Researcher in the Operations Research and Financial Engineering department and at the Bendheim Center for Finance at Princeton University.

Title: Nonlinear regression with nonstationarity and heteroscedasticity

Speaker: Qiying Wang, University of Sydney

Time: Thursday,November 28, 15:00-16:00

Place: Room 217, Guanghua Building 2

Abstract:

This paper develops an asymptotic theory of nonlinear least squares estimation by establishing a new framework that can be easily applied to various nonlinear regression models with heteroscedasticity. This paper explores an application of the framework to nonlinear regression models with nonstationarity and heteroscedasticity. In addition to these main results, this paper provides a maximum inequality for a class of martingales and establishes some new results on convergence to a local time and convergence to a mixture of normal distributions.

Introduction:

Qiying Wang is a Professor of Statistics and Econometrics at the University of Sydney, Australia, with well-established expertise in classical limit theory and asymptotics. He works on nonstationary time series econometrics, nonparametric statistics, econometric theory, martingale limit theory and self-normalized limit theory. He was an Australian Research Fellow from 2007 to 2012 and his research has been constantly supported by Australian Research Council since 2004. He has published over 80 research papers and over half of them appeared in the top ranked Probability, Statistics and Econometrics journals such as Econometrica, Annals of Probability, Annals of Statistics, Journal of Econometrics and Econometric Theory. He is one of the three recipients of 2017 Econometric Theory Plura Scripsit Award. His monograph entitled "Limit theorems for nonlinear cointegrating regression" systematically introduces the machinery of theoretical development in nonlinear cointegrating regression.

Your participation is warmly welcomed!

分享

010-62747206

北京大学光华管理学院2号楼

©2017 北京大学光华管理学院 版权所有 京ICP备05065075-1