首页 | 本学科首页   官方微博 | 高级检索  
     检索      


Prediction intervals for time series models with trend via sieve bootstrap
Authors:Grzegorz Chłapiński  Roman Różański
Institution:Institute of Mathematics and Computer Science, Wroc?aw University of Technology, Wroclaw, Poland
Abstract:This paper discusses method for constructing the prediction intervals for time series model with trend using the sieve bootstrap procedure. Gasser–Müller type of kernel estimator is used for trend estimation and prediction. The boundary modification of the kernel is applied to control the edge effect and to construct the predictor of a trend.
Keywords:Sieve bootstrap  Hybrid bootstrap  Bootstrap-t  Prediction intervals  Kernel estimator  Boundary kernel
本文献已被 ScienceDirect 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号