-
首页
-
- 首页
- WORKSHOP预告
计量、金融与大数据分析workshop：Machine Learning using Nonstationary Data
发布日期：2023-11-17 12:00 来源：
题目：Machine Learning using Nonstationary Data
摘要：Machine learning offers a promising set of tools for forecasting. However, some of the well-known properties do not apply to nonstationary data. This paper uses a simple procedure to extend machine learning methods to nonstationary data that does not require the researcher to have prior knowledge of which variables are nonstationary or the nature of the nonstationarity. I illustrate theoretically that using this procedure with LASSO or adaptive LASSO generates consistent variable selection on a mix of stationary and nonstationary explanatory variables. In an empirical exercise, I examine the success of this approach at forecasting U.S. inflation rates and the industrial production index using a number of different machine learning methods. I find that the proposed method either significantly improves prediction accuracy over traditional practices or delivers comparable performance, making it a reliable choice for obtaining stationary components of high-dimensional data.
主讲人：Jin Xi, University of California San Diego
主持人：（国发院）黄卓、张俊妮、孙振庭
（北大新结构经济学研究院）胡博
（经济学院）王一鸣、王熙、刘蕴霆、王法
时间：2023年11月17日（周五）10:00 AM -- 11:30 AM
地点：北京大学经济学院107
主讲人简介：
Jin Xi is a Ph.D. candidate from University of California San Diego. Her research interests include High-Dimensional Econometrics and Factor Models. She has publication in Social Choice and Welfare.
国家发展研究院官方微信
Copyright© 1994-2012 北京大学 国家发展研究院 版权所有, 京ICP备05065075号-1
保留所有权利，不经允许请勿挪用
-
首页
-
计量、金融与大数据分析workshop：Machine Learning using Nonstationary Data
发布日期：2023-11-17 12:00 来源：
题目：Machine Learning using Nonstationary Data
摘要：Machine learning offers a promising set of tools for forecasting. However, some of the well-known properties do not apply to nonstationary data. This paper uses a simple procedure to extend machine learning methods to nonstationary data that does not require the researcher to have prior knowledge of which variables are nonstationary or the nature of the nonstationarity. I illustrate theoretically that using this procedure with LASSO or adaptive LASSO generates consistent variable selection on a mix of stationary and nonstationary explanatory variables. In an empirical exercise, I examine the success of this approach at forecasting U.S. inflation rates and the industrial production index using a number of different machine learning methods. I find that the proposed method either significantly improves prediction accuracy over traditional practices or delivers comparable performance, making it a reliable choice for obtaining stationary components of high-dimensional data.
主讲人：Jin Xi, University of California San Diego
主持人：（国发院）黄卓、张俊妮、孙振庭
（北大新结构经济学研究院）胡博
（经济学院）王一鸣、王熙、刘蕴霆、王法
时间：2023年11月17日（周五）10:00 AM -- 11:30 AM
地点：北京大学经济学院107
主讲人简介：
Jin Xi is a Ph.D. candidate from University of California San Diego. Her research interests include High-Dimensional Econometrics and Factor Models. She has publication in Social Choice and Welfare.
访问学者申请
EMAIL
北京大学
ENGLISH
