Testing structure of the dispersion matrix

来源 :The 24th International Workshop on Matrices and Statistics(第 | 被引量 : 0次 | 上传用户:zdh313
下载到本地 , 更方便阅读
声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架
论文部分内容阅读
  Assume we have a sample of size n from a p-dimensional population with first four finite moments. We are interested in testing some basic hypothesis about the covariance structure in the dispersion matrix Σ. We are concentrating to the sphericity test and the uncorrelatedness test in the situation when both, the sample size and the number of variables can be large. As motivated in Srivastava (2005) we can not rely on maximum likelihood tests in this framework. For the test of sphericity we derive asymptotic expansion of a teststatistic constructed via trace functions of the sample dispersion matrix and find its asymptotic normal distribution under null-hypothesis. For the uncorrelatedness test a chi-square test-statistic is constructed. In a simulation study probabilistic behaviour of these test-statistics is studied and speed of convergence to the asymptotic distributions examined.
其他文献
  Many happy returns, Simo! To celebrate over 25 years of collaboration, we present an indexed and illustrated bibliography on the occasion of your 70th birth
  For a straight line in 2D, it is well known that the Total Least-Squares (TLS) estimator within an Errors-In-Variables (EIV) model generates the same soluti
会议
  Estimation of the covariance structure for irregular sparse longitudinal data has been studied by many authors in recent years but typically using fully par
会议
  Our motivation in this talk is the 13th-century Anxi iron-plate doubly-classic 6x 6 bordered magic square discussed by Kai-Tai Fang at the 22nd Internationa
  In this talk, we consider various Bayes factor approaches for the hypothesis testing problem in analysis-of-variance (ANOVA) designs. We firstly reparameter
会议
  With a history of more than 3000 years, magic squares still are mysterious in various aspects. We in this paper give a comprehensive review and study on cla
会议
  Research on quasi stationary distributions , is a very important research topic. When the sample size is not sufficiently large, the asymptotic results of t
  Integer-valued generalized autoregressive conditional heteroscedasticity (GARCH) models have played an important role in time series analysis of count data.
  We propose Partial Correlation Screening (PCS) as a new row-by-row approach. To estimate the i-th row of Ω, 1 ≤ i ≤ p, PCS uses a Screen step and a Clean
会议
  Methods of survival analysis (e.g., the Cox proportional hazards model) require that the event time be measured with respect to some origin time. The choice
会议