A Scalable Frequentist Model Averaging Method

doi:10.1080/07350015.2022.2116442

CSpace

	A Scalable Frequentist Model Averaging Method
	Zhu, Rong 1; Wang, Haiying 2; Zhang, Xinyu 3; Liang, Hua 4
	2022-09-26
发表期刊	JOURNAL OF BUSINESS & ECONOMIC STATISTICS
ISSN	0735-0015
页码	10
摘要	Frequentist model averaging is an effective technique to handle model uncertainty. However, calculation of the weights for averaging is extremely difficult, if not impossible, even when the dimension of the predictor vector, p, is moderate, because we may have 2(p) candidate models. The exponential size of the candidate model set makes it difficult to estimate all candidate models, and brings additional numeric errors when calculating the weights. This article proposes a scalable frequentist model averaging method, which is statistically and computationally efficient, to overcome this problem by transforming the original model using the singular value decomposition. The method enables us to find the optimal weights by considering at most p candidate models. We prove that the minimum loss of the scalable model averaging estimator is asymptotically equal to that of the traditional model averaging estimator. We apply the Mallows and Jackknife criteria to the scalable model averaging estimator and prove that they are asymptotically optimal estimators. We further extend the method to the high-dimensional case (i.e., p >= n). Numerical studies illustrate the superiority of the proposed method in terms of both statistical efficiency and computational cost.
关键词	Asymptotic optimality High-dimensional data Jackknife criterion Mallows criterion Singular value decomposition
DOI	10.1080/07350015.2022.2116442
收录类别	SCI
语种	英语
资助项目	NNSF of China[11301514] ; NNSF of China[71532013] ; Shanghai Municipal Science and Technology Major Project[2018SHZDZX01] ; 111 Project[B18015] ; NNSF grant[71925007] ; NNSF grant[72091212] ; NNSF grant[12288201] ; CAS Project for Young Scientists in Basic Research[YSBR-008] ; NSF[2105571]
WOS研究方向	Business & Economics ; Mathematical Methods In Social Sciences ; Mathematics
WOS类目	Economics ; Social Sciences, Mathematical Methods ; Statistics & Probability
WOS记录号	WOS:000863731800001
出版者	TAYLOR & FRANCIS INC
引用统计
文献类型	期刊论文
条目标识符	http://ir.amss.ac.cn/handle/2S8OKBNM/60842
专题	中国科学院数学与系统科学研究院
通讯作者	Liang, Hua
作者单位	1.Fudan Univ, Inst Sci & Technol Brain Inspired Intelligence, Shanghai, Peoples R China 2.Univ Connecticut, Dept Stat, Storrs, CT 06269 USA 3.Chinese Acad Sci, Acad Math & Syst Sci, Beijing, Peoples R China 4.George Washington Univ, Dept Stat, Washington, DC 20052 USA
推荐引用方式 GB/T 7714	Zhu, Rong,Wang, Haiying,Zhang, Xinyu,et al. A Scalable Frequentist Model Averaging Method[J]. JOURNAL OF BUSINESS & ECONOMIC STATISTICS,2022:10.
APA	Zhu, Rong,Wang, Haiying,Zhang, Xinyu,&Liang, Hua.(2022).A Scalable Frequentist Model Averaging Method.JOURNAL OF BUSINESS & ECONOMIC STATISTICS,10.
MLA	Zhu, Rong,et al."A Scalable Frequentist Model Averaging Method".JOURNAL OF BUSINESS & ECONOMIC STATISTICS (2022):10.