首页   按字顺浏览 期刊浏览 卷期浏览 A simulation study of l1:estimation of a seasonal moving average time series model
A simulation study of l1:estimation of a seasonal moving average time series model

 

作者: William T.M. Dunsmuir,  

 

期刊: Communications in Statistics - Simulation and Computation  （Taylor Available online 1992）
卷期: Volume 21, issue 2  

页码: 519-531

 

ISSN:0361-0918

 

年代: 1992

 

DOI:10.1080/03610919208813033

 

出版商: Marcel Dekker, Inc.

 

关键词: least absolute deviation estimation;seasonal moving average models;simulation in time series

 

数据来源: Taylor

 

摘要:

A detailed simulation study is reported on the application of l1:estimations to a seasonal moving average model. It is found that the asymptotic normal distribution is a nonapproximation to the finite sample distribution. However, the expected benefits of l1:estimation relative to l2:are partially realised for nonnormal innovative distributions.

 

点击下载:  PDF (319KB)



返 回