서지주요정보
Higher moments of matrix quadratic form = 행렬의 이차형에 대한 고차의 적률
서명 / 저자 Higher moments of matrix quadratic form = 행렬의 이차형에 대한 고차의 적률 / Chul Kang.
저자명 Kang, Chul ; 강철
발행사항 [대전 : 한국과학기술원, 1996].
Online Access 원문보기 원문인쇄

소장정보

등록번호

8011334

소장위치/청구기호

학술문화관(문화관) 보존서고

DMA 96010

휴대폰 전송

도서상태

이용가능

대출가능

반납예정일

초록정보

The importance of multivariate analysis in statistics has been recognized owing to its wide range of applications. And the matrix quadratic form plays a central role in multivariate analysis. The matrix quadratic form frequently appears in multivariate regression analysis in statistical inferrence, multivariate analysis of variance and test of statistical hypotheses, and has been much studied since 1980's. The moment problem of matrix quadratic form has attracted the attentions of many researchers. The theory of matrix quadratic form, however, leaves much to be desired because, in general, the enormous size of the matrix to be considered often complicates computations, and in some cases makes it impossible to compute. When a random matrix X is distributed the multivariate normal, XAX' is called the matrix quadratic form, where A is a constant matrix. In 1987, Neudecker and Wansbeek found the second moment of matrix quadratic form. In 1988 von Rosen solved the problem of any higher moment of the random matrix X, and as an application found the second moment of the matrix quadratic form by a method different from that of Neudecker and Wansbeek (1987). Using the result of von Rosen (1988), Tracy and Sultan found the third moment of matrix quadratic form in 1993. In this paper we derive some properties of the Kronecker product, vec operator and commutation matrix which are important tools in the moments of matrix quadratic form, and find any higher moment of matrix quadratic form by applying these properties and the result of von Rosen (1988). Morever, as an application of the general form of the higher moment of the matrix quadratic form we calculate any higher moment of the non-central Wishart distribution. Also, we provide various examples illustrating ours results. The results of this paper reveal novel formulae of matrix algebra and may shed light on the unsolvable problems.

서지기타정보

서지기타정보
청구기호 {DMA 96010
형태사항 iii, 55 p. : 삽도 ; 26 cm
언어 영어
일반주기 저자명의 한글표기 : 강철
지도교수의 영문표기 : Byung-Chun Kim
지도교수의 한글표기 : 김병천
학위논문 학위논문(박사) - 한국과학기술원 : 수학과,
서지주기 Reference : p. 54-55
주제 Kronecker product
Commutation matrix
Vec operator
Moments of matrix
Matrix quadratic form
Non-central wishart distribution
Kronecker 곱
교환 행렬
Vec 연산
행렬의 적률
행렬의 이차형
비중심 Wishart 분포
QR CODE qr code