• Proceedings of the Korean Statistical Society Conference
• /
• 2002.05a
• /
• pp.183-188
• /
• 2002

Multiple imputation, proposed by Rubin, is a procedure for handling missing data. One of the attractive parts of multiple imputation is the simplicity of the variance estimation formula. Because of the simplicity, it has been often abused and misused beyond its original prescription. This paper provides the bias of the multiple imputation variance estimator for a linear point estimator and discusses when the bias can be safely neglected.

• Communications for Statistical Applications and Methods
• /
• v.10 no.2
• /
• pp.361-370
• /
• 2003

Those who are interested in making inferences concerning linear combination of valiance components in a simple linear regression model with unbalanced nested error structure can use the confidence intervals proposed in this paper. Two approximate confidence intervals for the sum of two variance components in the model are proposed. Simulation study is peformed to compare the methods. The methods are applied to a numerical example and recommendations are given for choosing a proper interval.

• Communications for Statistical Applications and Methods
• /
• v.13 no.3
• /
• pp.657-667
• /
• 2006

The concept of entropy, introduced in communication theory by Shannon (1948) as a measure of uncertainty, is of prime interest in information-theoretic statistics. This paper considers the minimum variance unbiased estimation for the maximum entropy of the transformed inverse Gaussian random variable by $Y=X^{-1/2}$. The properties of the derived UMVU estimator is investigated.

• Proceedings of the Korean Statistical Society Conference
• /
• 2002.11a
• /
• pp.327-332
• /
• 2002

Variance estimation for the regression estimator for a two-phase sample is investigated. A replication variance estimator with number of replicates equal to or slightly larger than the size of the second-phase sample is developed. In these cases, the proposed method is asymptotically equivalent to the full jackknife, but uses smaller number of replications.

• Communications for Statistical Applications and Methods
• /
• v.4 no.1
• /
• pp.111-117
• /
• 1997

The ordinary least squares estimator $S^2$ for the variance of the disturbances is considered in the linear regression model with sutocorrelated disturbances. It is proved that the OLS-estimator of disturbance variance is asymptotically unbiased and weakly consistent, when the distrubances are generated by an MA(q) process. In particular, the asymptotic unbiasedness and consistency of $S^2$ is satisfied without any restriction on the regressor matrix.

• Communications for Statistical Applications and Methods
• /
• v.6 no.2
• /
• pp.487-500
• /
• 1999

We propose a class of hierarchical Bayes estimators of the error variance under the relative squared error loss in balanced fixed-effects two-way analysis of variance models. Also we provide analytic expressions for the risk improvement of the hierarchical Bayes estimators over multiples of the error sum of squares. Using these expressions we identify a subclass of the hierarchical Bayes estimators each member of which dominates the best multiple of the error sum of squares which is known to be minimax. Numerical values of the percentage risk improvement are given in some special cases.

• Communications for Statistical Applications and Methods
• /
• v.20 no.4
• /
• pp.329-337
• /
• 2013

Consider a p-variate($p{\geq}3$) normal distribution with mean ${\theta}$ and covariance matrix ${\sum}={\sigma}^2{\mathbf{I}}_p$ for any unknown scalar ${\sigma}^2$. In this paper we improve the James-Stein estimator of ${\theta}$ in cases of shrinking toward some vectors using the Stein variance estimator. It is also shown that this domination does not hold for the positive part versions of these estimators.

• Communications for Statistical Applications and Methods
• /
• v.9 no.2
• /
• pp.459-471
• /
• 2002

In applications using a linear regression model with nested error structure, one might be interested in making inferences concerning variance components. This article proposes approximate confidence intervals on the variance component of the primary level in a simple linear regression model with an unbalanced nested error structure. The intervals are compared using computer simulation and recommendations are provided for selecting an appropriate interval.

• 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.

• Communications for Statistical Applications and Methods
• /
• v.9 no.1
• /
• pp.249-259
• /
• 2002

In order to estimate the mean and variance for a Normal distribution which is truncated at both right and left sides, maximum likelihood estimators based on the entire sample from the original distribution are compared with the sample mean and variance of the censored sample which is the data remaining after truncation using simulation. We found that, surprisingly, the mean squared error of the mean based on the censored data Is smaller than that of the full sample estimators.