• Journal of the Korean Statistical Society
• /
• v.15 no.2
• /
• pp.97-106
• /
• 1986

Assume that $X_1, X_2, \cdots, X_N$ are iid p-dimensional normal random vectors ($p \geq 3$) with unknown covariance matrix. The problem of estimating multivariate normal mean vector in an empirical Bayes situation is considered. Empirical Bayes estimators, obtained by Bayes treatmetn of the covariance matrix, are presented. It is shown that the estimators are minimax, each of which domainates teh maximum likelihood estimator (MLE), when the loss is nonsingular quadratic loss. We also derive approximate credibility region for the mean vector that takes advantage of the fact that the MLE is not the best estimator.

• Journal of the Korean Mathematical Society
• /
• v.42 no.5
• /
• pp.959-973
• /
• 2005

In this paper, based on large deviation probabilities on Gaussian random vectors, we obtain Strassen's functional LIL for d-dimensional self-similar Gaussian process in Holder norm via estimating large deviation probabilities for d-dimensional self-similar Gaussian process in Holder norm.

• Communications for Statistical Applications and Methods
• /
• v.12 no.3
• /
• pp.635-642
• /
• 2005

In statistical diagnostics a large number of influence measures have been proposed for identifying outliers and influential observations. However it seems to be few accounts of the influence diagnostics on test statistics. We study influence analysis on the likelihood ratio test statistic whether the two sets of variables are uncorrelated with one another or not. The influence of observations is measured using the case-deletion approach, the influence function. We compared the proposed influence measures through two illustrative examples.

• Journal of the Korean Statistical Society
• /
• v.13 no.1
• /
• pp.20-24
• /
• 1984

In analogy to the theorem proved by So and Jeon (1982), we give a multi-dimensional version of local limit theorem for large deviations in the continuous case. We also prove a similar theorem in the case of lattice random vectors. Some examples are given.

• International Journal of Reliability and Applications
• /
• v.7 no.2
• /
• pp.127-140
• /
• 2006

AIn this paper we introduce and study a multivariate notions of mean inactivity time (MIT) functions. Basic properties of these functions are derived and their relationship to the multivariate conditional reversed hazard rate functions is studied. A partial ordering, called MIT ordering, of non-negative random vectors is introduced and its basic properties are presented. Its relationship to reversed hazard rate ordering is pointed out. Finally, using the MIT ordering, a bivariate and multivariate notions of IMIT (increasing mean inactivity time) class is introduced and studied.

• 한국데이터정보과학회:학술대회논문집
• /
• 2001.10a
• /
• pp.13-16
• /
• 2001

We compare methods for detecting influential observations that have a large influence on the likelihood ratio test statistics that the two sets of variables are uncorrelated with one another. For this purpose we derive results of the deletion diagnostic, the influence function, the standardized influence matrix and the local influence. An illustrative example is given.

• Journal of Korean Society for Quality Management
• /
• v.8 no.1
• /
• pp.8-15
• /
• 1980

There are many methods of pattern recognition. In this paper we assume that the responses of independent m groups are described by p-variate normal random variables with distinct mean vectors and a common covariance matrix. Under the assumption we give pattern recognition of m groups by means of canonical discrininant analysis and it's computer program. An example is presented.

• Communications for Statistical Applications and Methods
• /
• v.27 no.1
• /
• pp.149-161
• /
• 2020

In this paper we extend the functional principal component analysis for real-valued random functions to the case of Hilbert-space-valued functional random objects. For this, we introduce an autocovariance operator acting on the space of real-valued functions. We establish an eigendecomposition of the autocovariance operator and a Karuhnen-Loève expansion. We propose the estimators of the eigenfunctions and the functional principal component scores, and investigate the rates of convergence of the estimators to their targets. We detail the implementation of the methodology for the cases of compositional vectors and density functions, and illustrate the method by analyzing time-varying population composition data. We also discuss an extension of the methodology to multivariate cases and develop the corresponding theory.

• Journal of applied mathematics & informatics
• /
• v.18 no.1_2
• /
• pp.247-259
• /
• 2005

Necessary and sufficient conditions, under which there exists (at least) a sequence of vectors of real numbers for which the distribution function (d.f.) of any vector of extreme order statistics converges to a non-degenerate limit, are derived. The interesting thing is that these conditions solely depend on the univariate marginals. Moreover, the limit splits into the product of the limit univariate marginals if all the bivariate marginals of the trivariate d.f., from which the sample is drawn, is of negative quadrant dependent random variables (r.v.'s). Finally, all these results are stated for the multivariate extremes with arbitrary dimensions.

• Journal of the Korean Statistical Society
• /
• v.25 no.2
• /
• pp.243-253
• /
• 1996

Let X and Y be random vectors in R$^{n}$ . A random vector X is 'more associated' than Y if and only if P(X $\in$ A ∩ B) - P(X $\in$ A)P(X $\in$ B) $\geq$ P(Y $\in$ A ∩ B)-P(Y $\in$ A)P(Y $\in$ B) for all open upper sets A and B. By requiring the above inequality to hold for some open upper sets A and B various notions of positive dependence orderings which are weaker than 'more associated' ordering are obtained. First a general theory is given and then the results are specialized to some concepts of a particular interest. Various properties and interrelationships are derived.