The multivariate split normal distribution and asymmetric principal components analysis

被引:20
|
作者
Villani, M
Larsson, R
机构
[1] Stockholm Univ, Sveriges Riksbank, Div Res, SE-10337 Stockholm, Sweden
[2] Stockholm Univ, Dept Stat, SE-10337 Stockholm, Sweden
[3] Uppsala Univ, Dept Informat Sci, Uppsala, Sweden
关键词
Bayesian inference; estimation; maximum likelihood; multivariate analysis; skewness; statistical distribution;
D O I
10.1080/03610920600672252
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
The multivariate split normal distribution extends the usual multivariate normal distribution by a set of parameters which allows for skewness in the form of contraction/dilation along a subset of the principal axes. This article derives some properties for this distribution, including its moment generating function, multivariate skewness, and kurtosis, and discusses its role as a population model for asymmetric principal components analysis. Maximum likelihood estimators and a complete Bayesian analysis, including inference on the number of skewed dimensions and their directions, are presented.
引用
收藏
页码:1123 / 1140
页数:18
相关论文
共 50 条
  • [1] A dynamic principal components analysis based on multivariate matrix normal dynamic linear models
    Salvador, M
    Gallizo, JL
    Gargallo, P
    JOURNAL OF FORECASTING, 2003, 22 (6-7) : 457 - 478
  • [2] Multivariate Taguchi loss function optimization based on principal components analysis and normal boundary intersection
    Fabrício Alves de Almeida
    Ana Carolina Oliveira Santos
    Anderson Paulo de Paiva
    Guilherme Ferreira Gomes
    José Henrique de Freitas Gomes
    Engineering with Computers, 2022, 38 : 1627 - 1643
  • [3] Multivariate Taguchi loss function optimization based on principal components analysis and normal boundary intersection
    de Almeida, Fabricio Alves
    Oliveira Santos, Ana Carolina
    de Paiva, Anderson Paulo
    Gomes, Guilherme Ferreira
    de Freitas Gomes, Jose Henrique
    ENGINEERING WITH COMPUTERS, 2022, 38 (02) : 1627 - 1643
  • [4] Multivariate prediction with nonlinear principal components analysis: Application
    Blasius, J
    Gower, JC
    QUALITY & QUANTITY, 2005, 39 (04) : 373 - 390
  • [5] Multivariate Prediction with Nonlinear Principal Components Analysis: Application
    JÖRG BLASIUS
    JOHN C. GOWER
    Quality and Quantity, 2005, 39 : 373 - 390
  • [6] Multivariate Prediction with Nonlinear Principal Components Analysis: Theory
    JOHN C. GOWER
    JÖRG BLASIUS
    Quality and Quantity, 2005, 39 : 359 - 372
  • [7] Multivariate prediction with nonlinear principal components analysis: Theory
    Gowen, JC
    Blasius, J
    QUALITY & QUANTITY, 2005, 39 (04) : 359 - 372
  • [8] Visualizing principal components analysis for multivariate process data
    Bisgaard, Soren
    Huang, Xuan
    JOURNAL OF QUALITY TECHNOLOGY, 2008, 40 (03) : 299 - 309
  • [9] MULTIVARIATE ALLOMETRY AND REMOVAL OF SIZE WITH PRINCIPAL COMPONENTS-ANALYSIS
    SOMERS, KM
    SYSTEMATIC ZOOLOGY, 1986, 35 (03): : 359 - 368
  • [10] METHOD OF PRINCIPAL COMPONENTS IN MULTIVARIATE STATISTICAL-ANALYSIS (REVIEW)
    ERMAKOV, SM
    POKHODZEI, BB
    SOLNTSEV, VN
    INDUSTRIAL LABORATORY, 1983, 49 (05): : 498 - 508