A new model selection procedure based on dynamic quantile regression

doi:10.1080/02664763.2014.909787

	A new model selection procedure based on dynamic quantile regression
	Xiong, Wei ; Tian, Maozai
	2014-10
发表期刊	JOURNAL OF APPLIED STATISTICS
卷号	41 期号:10 页码:2240-2256
摘要	In this article, we propose a novel robust data-analytic procedure, dynamic quantile regression (DQR), for model selection. It is robust in the sense that it can simultaneously estimate the coefficients and the distribution of errors over a large collection of error distributions even those that are heavy-tailed and may not even possess variances or means; and DQR is easy to implement in the sense that it does not need to decide in advance which quantile(s) should be gathered. Asymptotic properties of related estimators are derived. Simulations and illustrative real examples are also given.
关键词	dynamic quantile regression variable selection error distribution DQR-oracle asymptotic relative efficiency heteroskedasticity
DOI	10.1080/02664763.2014.909787
收录类别	SCI ; SCOPUS ; SCIE
ISSN	0266-4763
语种	英语
WOS研究方向	Mathematics
WOS类目	Statistics & Probability
WOS记录号	WOS:000337959200010
出版者	TAYLOR & FRANCIS LTD
原始文献类型	Article
EISSN	1360-0532
引用统计	被引频次：2[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.lzufe.edu.cn/handle/39EH0E1M/1925
专题	统计与数据科学学院
作者单位	1.Renmin Univ China, Sch Stat, Ctr Appl Stat, Beijing 100872, Peoples R China; 2.Lanzhou Univ Finance & Econ, Sch Stat, Lanzhou 730101, Gansu, Peoples R China
推荐引用方式 GB/T 7714	Xiong, Wei,Tian, Maozai. A new model selection procedure based on dynamic quantile regression[J]. JOURNAL OF APPLIED STATISTICS,2014,41(10):2240-2256.
APA	Xiong, Wei,&Tian, Maozai.(2014).A new model selection procedure based on dynamic quantile regression.JOURNAL OF APPLIED STATISTICS,41(10),2240-2256.
MLA	Xiong, Wei,et al."A new model selection procedure based on dynamic quantile regression".JOURNAL OF APPLIED STATISTICS 41.10(2014):2240-2256.