• Title/Summary/Keyword: multivariate normal populations

Search Result 19, Processing Time 0.02 seconds

A BAYESIAN METHOD FOR FINDING MINIMUM GENERALIZED VARIANCE AMONG K MULTIVARIATE NORMAL POPULATIONS

  • Kim, Hea-Jung
    • Journal of the Korean Statistical Society
    • /
    • v.32 no.4
    • /
    • pp.411-423
    • /
    • 2003
  • In this paper we develop a method for calculating a probability that a particular generalized variance is the smallest of all the K multivariate normal generalized variances. The method gives a way of comparing K multivariate populations in terms of their dispersion or spread, because the generalized variance is a scalar measure of the overall multivariate scatter. Fully parametric frequentist approach for the probability is intractable and thus a Bayesian method is pursued using a variant of weighted Monte Carlo (WMC) sampling based approach. Necessary theory involved in the method and computation is provided.

On the Robustness of Chi-square Test Procedure for a Compounded Multivariate Normal Mean

  • Kim, Hea-Jung
    • Communications for Statistical Applications and Methods
    • /
    • v.2 no.2
    • /
    • pp.330-335
    • /
    • 1995
  • The rebustness of one sample Chi-square test for multivariate normal mean vector is investigated when the multivariate normal population is mixed with another multivariate normal population with differing in the mean vector. Explicit expressions for the level of significance and power of the test are derived. Some numerical results indicate that the Chi-square test procedure is quite robust against slight mixtures of multivariate normal populations differing in location parameters.

  • PDF

A Bayesian Criterion for a Multiple test of Two Multivariate Normal Populations

  • Kim Hea-Jung;Son Young Sook
    • Proceedings of the Korean Statistical Society Conference
    • /
    • 2000.11a
    • /
    • pp.147-152
    • /
    • 2000
  • A Bayesian criterion is proposed for a multiple test of two independent multivariate normal populations. For a Bayesian test the fractional Bayes facto.(FBF) of O'Hagan(1995) is used under the assumption of Jeffreys priors, noninformative improper proirs. In this test the FBF without the need of sampling minimal training samples is much simpler to use than the intrinsic Bayes facotr(IBF) of Berger and Pericchi(1996). Finally, a simulation study is performed to show the behaviors of the FBF.

  • PDF

A Bayesian Criterion for a Multiple test of Two Multivariate Normal Populations

  • Kim, Hae-Jung;Son, Young-Sook
    • Communications for Statistical Applications and Methods
    • /
    • v.8 no.1
    • /
    • pp.97-107
    • /
    • 2001
  • A simultaneous test criterion for multiple hypotheses concerning comparison of two multivariate normal populations is considered by using the so called Bayes factor method. Fully parametric frequentist approach for the test is not available and thus Bayesian criterion is pursued using a Bayes factor that eliminates its arbitrariness problem induced by improper priors. Specifically, the fractional Bayes factor (FBF) by O'Hagan (1995) is used to derive the criterion. Necessary theories involved in the derivation an computation of the criterion are provided. Finally, an illustrative simulation study is given to show the properties of the criterion.

  • PDF

On Testing Equality of Matrix Intraclass Covariance Matrices of $K$Multivariate Normal Populations

  • Kim, Hea-Jung
    • Communications for Statistical Applications and Methods
    • /
    • v.7 no.1
    • /
    • pp.55-64
    • /
    • 2000
  • We propose a criterion for testing homogeneity of matrix intraclass covariance matrices of K multivariate normal populations, It is based on a variable transformation intended to propose and develop a likelihood ratio criterion that makes use of properties of eigen structures of the matrix intraclass covariance matrices. The criterion then leads to a simple test that uses an asymptotic distribution obtained from Box's (1949) theorem for the general asymptotic expansion of random variables.

  • PDF

Testing Homogeneity of Diagonal Covariance Matrices of K Multivariate Normal Populations

  • Kim, Hea-Jung
    • Communications for Statistical Applications and Methods
    • /
    • v.6 no.3
    • /
    • pp.929-938
    • /
    • 1999
  • We propose a criterion for testing homogeneity of diagonal covariance matrices of K multivariate normal populations. It is based on a factorization of usual likelihood ratio intended to propose and develop a criterion that makes use of properties of structures of the diagonal convariance matrices. The criterion then leads to a simple test as well as to an accurate asymptotic distribution of the test statistic via general result by Box (1949).

  • PDF

A Bayesian Comparison of Two Multivariate Normal Genralized Variances

  • Kim, Hea-Jung
    • Proceedings of the Korean Statistical Society Conference
    • /
    • 2002.05a
    • /
    • pp.73-78
    • /
    • 2002
  • In this paper we develop a method for constructing a Bayesian HPD (highest probability density) interval of a ratio of two multivariate normal generalized variances. The method gives a way of comparing two multivariate populations in terms of their dispersion or spread, because the generalized variance is a scalar measure of the overall multivariate scatter. Fully parametric frequentist approaches for the interval is intractable and thus a Bayesian HPD(highest probability densith) interval is pursued using a variant of weighted Monte Carlo (WMC) sampling based approach introduced by Chen and Shao(1999). Necessary theory involved in the method and computation is provided.

  • PDF

Bootstrap Confidence Intervals of Classification Error Rate for a Block of Missing Observations

  • Chung, Hie-Choon
    • Communications for Statistical Applications and Methods
    • /
    • v.16 no.4
    • /
    • pp.675-686
    • /
    • 2009
  • In this paper, it will be assumed that there are two distinct populations which are multivariate normal with equal covariance matrix. We also assume that the two populations are equally likely and the costs of misclassification are equal. The classification rule depends on the situation when the training samples include missing values or not. We consider the bootstrap confidence intervals for classification error rate when a block of observation is missing.

Discriminant Analysis with Icomplete Pattern Vectors

  • Hie Choon Chung
    • Communications for Statistical Applications and Methods
    • /
    • v.4 no.1
    • /
    • pp.49-63
    • /
    • 1997
  • We consider the problem of classifying a p x 1 observation into one of two multivariate normal populations when the training smaples contain a block of missing observation. A new classification procedure is proposed which is a linear combination of two discriminant functions, one based on the complete samples and the other on the incomplete samples. The new discriminant function is easy to use.

  • PDF

A VARIABLE SELECTION IN HETEROSCEDASTIC DISCRIVINANT ANALYSIS : GENERAL PREDICTIVE DISCRIMINATION CASE

  • Kim, Hea-Jung
    • Journal of the Korean Statistical Society
    • /
    • v.21 no.1
    • /
    • pp.1-13
    • /
    • 1992
  • This article deals with variable selection problem under a newly formed predictive heteroscedastic discriminant rule that accounts for mulitple homogeneous covariance matrices across the K multivariate normal populations. A general version of predictive discriminant rule, a variable selection criterion, and a criterion for stopping with further selection are suggested. In a simulation study the practical utilities of those considered are demonstrated.

  • PDF