# [1]温阳俊,朱道元.一类新的多元t-型分布[J].东南大学学报(自然科学版),2007,37(3):543-548.[doi:10.3969/j.issn.1001-0505.2007.03.038] 　Wen Yangjun,Zhu Daoyuan.New class of multivariate t-type distributions[J].Journal of Southeast University (Natural Science Edition),2007,37(3):543-548.[doi:10.3969/j.issn.1001-0505.2007.03.038] 点击复制 一类新的多元t-型分布() 分享到： var jiathis_config = { data_track_clickback: true };

37

2007年第3期

543-548

2007-05-20

## 文章信息/Info

Title:
New class of multivariate t-type distributions

1 东南大学理学院, 南京 210096; 2 南京农业大学数学系, 南京 210095
Author(s):
1 School of Sciences, Southeast University, Nanjing 210096, China
2 Department of Mathmatics, Nanjing Agricultural University, Nanjing 210095, China

Keywords:

O212.4
DOI:
10.3969/j.issn.1001-0505.2007.03.038

Abstract:
In order to found the sample theory based on the elliptically contoured distribution, the distributions of random vectors must be extended to that of random matrices. In this paper, definitions of matrix Kotz-type distribution, matrix inverse gamma distribution and matrix t-type distribution are proposed through three different methods, i.e., density generator approach, inverse Wishart distribution approach and construction approach of the matrix t-type distribution using two independent random matrices. And it is proved that they are densities of matrix distribution. The properties about matrix t-type distribution are stressed. Stochastic representation, expectation, linear transformations and quadratic forms are derived.

## 参考文献/References:

[1] Nadarajah S.Sampling distributions associated with the multivariate t distribution[J]. Statistica Neerlandica,2005,59(2):214-234.
[2] 朱道元,吴诚鸥,秦伟良.多元统计分析与软件SAS[M].南京:东南大学出版社,1999:174,175,197.
[3] Butler R J,McDonald J B,Nelson R D,et al.Robust and partially adaptive estimation of regression models[J]. Rev Econom Statist, 1990,72:321-327.
[4] Arslan O.Family of multivariate generalized t distributions[J].J Multivar Anal,2004,89(2):329-337.
[5] Arslan O.A new class of multivariate distribution:scale mixture of Kotz-type distributions[J]. Statist Prob Lett,2005,75(11):18-28.
[6] Johnson N L,Kotz S,Balakrishnan N.Continuous univariate distributions[M].New York:Wiley,1995:401-401.
[7] 方开泰,张尧庭.广义多元分析[M].北京:科学出版社,1993:122-145.