The rapid development and wide application of computer techniques permits to col-lect and store a huge amount data,where the number of measured variables is usuallylarge.But in high-dimensional case,many classical estimators and testing statistics be-have badly.For example,these estimators may significantly overestimate true parametersor the type I errors may tend to 1.These problems can not be accepted in statisticalresearches.Therefore,we have to establish the high dimensional properties of some clas-sical statistics.It is often-used to test whether the two population covariance matrices areproportional.So the purpose of this paper is to obtNn the high-dimensional properties ofthe statistics of the testing problem.... |