期刊论文详细信息
JOURNAL OF MULTIVARIATE ANALYSIS 卷:164
Asymptotic normality of quadratic forms with random vectors of increasing dimension
Article
Peng, Hanxiang1  Schick, Anton2 
[1] Indiana Univ Purdue Univ, Dept Math Sci, Indianapolis, IN 46202 USA
[2] SUNY Binghamton, Dept Math Sci, Binghamton, NY 13902 USA
关键词: Chi-square test with increasing number of cells;    Empirical likelihood;    Equal marginals;    Independence of components of high-dimensional normal random vectors;    Lindeberg condition;    Martingale central limit theorem;   
DOI  :  10.1016/j.jmva.2017.11.002
来源: Elsevier
PDF
【 摘 要 】

This paper provides sufficient conditions for the asymptotic normality of quadratic forms of averages of random vectors of increasing dimension and improves on conditions found in the literature. Such results are needed in applications of Owen's empirical likelihood when the number of constraints is allowed to grow with the sample size. Indeed, the results of this paper are already used in Peng and Schick (2013) for this purpose. We also demonstrate how our results can be used to obtain the asymptotic distribution of the empirical likelihood with an increasing number of constraints under contiguous alternatives. In addition, we discuss potential applications of our result. The first example focuses on a chi-square test with an increasing number of cells. The second example treats testing for the equality of the marginal distributions of a bivariate random vector. The third example generalizes a result of Schott (2005) by showing that a standardized version of his test for diagonality of the dispersion matrix of a normal random vector is asymptotically standard normal even if the dimension increases faster than the sample size. Schott's result requires the dimension and the sample size to be of the same order. (C) 2017 Elsevier Inc. All rights reserved.

【 授权许可】

Free   

【 预 览 】
附件列表
Files Size Format View
10_1016_j_jmva_2017_11_002.pdf 444KB PDF download
  文献评价指标  
  下载次数:18次 浏览次数:0次