Go to the main menu
Skip to content
Go to bottom
REFERENCE LINKING PLATFORM OF KOREA S&T JOURNALS
> Journal Vol & Issue
Korean Journal of Applied Statistics
Journal Basic Information
Journal DOI :
The Korean Statistical Society
Editor in Chief :
Volume & Issues
Volume 15, Issue 2 - Sep 2002
Volume 15, Issue 1 - Mar 2002
Selecting the target year
Categorizing tumor size as a prognostic factor for risk of relapse of hepatocellular carcinoma
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 1~8
DOI : 10.5351/KJAS.2002.15.1.001
Categorizing prognostic factors is very useful for a disease diagnosis, determination of treatment and study eligibility criteria. Methods often used to categorize factors are to select a cutpoint by biological theory, by graphical examination, by the minimum p-value approach. The last method involves multiple testing, and several methods for adjusting p-values have been developed. This study determines the cutpoint of tumor size to separate patients of high risk of relapse after hepatic resection of hepatocellular
On the Effects of English Emersion Program for School Students
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 9~20
DOI : 10.5351/KJAS.2002.15.1.009
This paper is to verify the effects of English immersion program through the case study of the three and an half week English camp run by Hongik University. The student's proficiency in English varies according to different factors. Another goal of this paper is to pin down certain objective factors that have an effect on the English proficiency and the improvement of the proficient. The generalized linear model(GLM) is adopted for the related analyses in this paper.
Statistical Outliers in Florida Counties at the Presidential Election 2000
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 21~32
DOI : 10.5351/KJAS.2002.15.1.021
We searched out in the votes data of the State of Florida at presidential election 2000. We used a multivariate regression analysis. We got there were several outliers including Palm Beach County. It means that we should analyze the number of disqualified ballots which were double-punched as well as the votes, to insist the " Butterfly Ballot" made Palm Beach outlier.
Comparing Accuracy of Imputation Methods for Categorical Incomplete Data
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 33~43
DOI : 10.5351/KJAS.2002.15.1.033
Various kinds of estimation methods have been developed for imputation of categorical missing data. They include category method, logistic regression, and association rule. In this study, we propose two fusions algorithms based on both neural network and voting scheme that combine the results of individual imputation methods. A Mont-Carlo simulation is used to compare the performance of these methods. Five factors used to simulate the missing data pattern are (1) input-output function, (2) data size, (3) noise of input-output function (4) proportion of missing data, and (5) pattern of missing data. Experimental study results indicate the following: when the data size is small and missing data proportion is large, modal category method, association rule, and neural network based fusion have better performances than the other methods. However, when the data size is small and correlation between input and missing output is strong, logistic regression and neural network barred fusion algorithm appear better than the others. When data size is large with low missing data proportion, a large noise, and strong correlation between input and missing output, neural networks based fusion algorithm turns out to be the best choice.
A Study of Composite Estimator in 2-level Rotation Design based on 3 Rotation Groups
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 45~55
DOI : 10.5351/KJAS.2002.15.1.045
The 2-level rotation design based on 3 rotation groups is discussed in view of Monthly Retail Trade Survey conducted by the Bureau of Census in U.S., and composite estimators for population characteristics are concerned. The generalized composite estimators and the recursive composite estimators are presented at 2-level rotation design with design gap and variance formulas for the composite estimators are provided. Also under the response variability related with covariance structure and correlation structure from repeated response, relative efficiencies of the composite estimators are compared.
Bayesian model selection in exponential survival models
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 57~71
DOI : 10.5351/KJAS.2002.15.1.057
We introduce three types of exponential survival models, such as simple model, change-point model and finite mixture model in this paper. Among these models, in order to choose the best model, the model choice method is proposed using Gelfand and Ghosh(1998)'s idea. Then to avoid the computational difficulties, data augmentation method (Tanner and Wong, 1987) and Gibbs sampler (Gelfand and Smith, 1990) are employed. Our methodology is applied to both simulated data and Stangl (1991)'s On-impramint Hydrochloride data.
Bayesian Variable Selection in Linear Regression Models with Inequality Constraints on the Coefficients
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 73~84
DOI : 10.5351/KJAS.2002.15.1.073
Linear regression models with inequality constraints on the coefficients are frequently used in economic models due to sign or order constraints on the coefficients. In this paper, we propose a Bayesian approach to selecting significant explanatory variables in linear regression models with inequality constraints on the coefficients. Bayesian variable selection requires computation of posterior probability of each candidate model. We propose a method which computes all the necessary posterior model probabilities simultaneously. In specific, we obtain posterior samples form the most general model via Gibbs sampling algorithm (Gelfand and Smith, 1990) and compute the posterior probabilities by using the samples. A real example is given to illustrate the method.
Quantification and Graphical Method for DNA Fingerprinting
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 85~105
DOI : 10.5351/KJAS.2002.15.1.085
To explore the relationships among frequencies for sets of alleles, within or between loci, is one of the first analyses in population genetic study. The general question is whether the frequency of a set of alleles is the same as the product of each of the separate allele frequencies. For two alleles of a single locus, Hardy-Weinberg equilibrium is tested and for an allele from each of two loci, linkage disequilibrium is tested. However, it is more useful if we can quantify and graphically represent this information. In this study, we suggest graphical methods to find associations between alleles. We also analyze the STR data of Korean population as an illustration.
Optimal Replacement Policy for a Repairable System with Combination Warranty
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 107~117
DOI : 10.5351/KJAS.2002.15.1.107
In this paper we present the optimal replacement policies following the expiration of combination warranty. We consider two types of combination warranty policies: renewing warranty and non-renewing warranty. The criterion used to determine the optimal replacement period is the expected cost rate per unit time from the user'perspective. The optimal maintenance period following the expiration of combination warranty is obtained. Some numerical examples are presented for illustrative purpose
Hierachical Bayes Estimation of Small Area Means in Repeated Survey
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 119~128
DOI : 10.5351/KJAS.2002.15.1.119
In this paper, we consider the HB estimators of small area means with repeated survey. mao and Yu(1994) considered small area model with repeated survey data and proposed empirical best linear unbiased estimators. We propose a hierachical Bayes version of Rao and Yu by assigning prior distributions for unknown hyperparameters. We illustrate our HB estimator using very popular data in small area problem and then compare the results with the estimator of Census Bureau and other estimators previously proposed.
A generalized logit model with mixed effects for categorical data
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 129~137
DOI : 10.5351/KJAS.2002.15.1.129
This paper suggests a generalized logit model with mixed effects for analysing frequency data in multi-contingency table. In this model nominal response variable is assumed to be polychotomous. When some factors are fixed but considered as ordinal and others are random, this paper shows how to use baseline-category logits to incoporate the mixed-effects of those factors into the model. A numerical algorithm was used to estimate model parameters by using marginal log-likelihood.
Bayesian Mode1 Selection and Diagnostics for Nonlinear Regression Model
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 139~151
DOI : 10.5351/KJAS.2002.15.1.139
This study is concerned with model selection and diagnostics for nonlinear regression model through Bayes factor. In this paper, we use informative prior and simulate observations from the posterior distribution via Markov chain Monte Carlo. We propose the Laplace approximation method and apply the Laplace-Metropolis estimator to solve the computational difficulty of Bayes factor.
Use of MonteCarlo Simulation for More Effective Teaching of Business Statistics
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 153~163
DOI : 10.5351/KJAS.2002.15.1.153
There seems to be a little trouble in delivering an effective teaching of statistics at the school of Business. Students attitude toward the understanding of mathematics appears to be one of the hindrances encountered to the wide spread use of statistics. Nevertheless, the usefulness of statistics is notably appreciated these days although students may not be aware of it. It is true that many processes of decision-making are based on various future uncertainties, however, decisions must be made at the present. In this synopsis, we would like to solve simple business problems, thereby, suggest more constructive ways to match the statistics with the interest of management. We, in the mean time, will talk about the understanding of business, method of teaching, and Monte Carlo education.
Development of Manpower Cultivation Index for Evaluation of Science and Technology Projects
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 165~177
DOI : 10.5351/KJAS.2002.15.1.165
This poper proposes a manpower cultivation index(MCI) which can be used for evaluating how much a research project contributes for manpower cultivation. The four factors which are considered to influence the manpower cultivation are the number of participated graduate students(research assistants) , the number of graduate students who obtain master or Ph. D. degrees among the participated graduate students in the research project, the number of research papers published by participated graduate students, and the number of graduate students who get jobs. By the proposed MCI, three major research projects of KOSEF are analysed in terms of manpower cultivation. These are SRC/ERC(Science research center/Engineering research center) and RRC(Regional research center) and OBR(Objective based research) . The computed MCls show that the three major research projects enjoy satisfactory manpower cultivation to some extent.
Measures for Evaluating Nearly Orthogonal Blocking
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 179~186
DOI : 10.5351/KJAS.2002.15.1.179
We usually execute the blocking under heterogeneity of experimental condition in response surface methodology. We can suggest the measure for evaluating nearly orthogonal blocking under second order models.
Investigation on Exact Tests
Korean Journal of Applied Statistics, volume 15, issue 1, 2002, Pages 187~199
DOI : 10.5351/KJAS.2002.15.1.187
When the sample size is small, exact tests are often employed because the asymptotic distribution of the test statistic is in doubt. The advantage of exact tests is that it is guaranteed to bound the type I error probability to the nominal level. In this paper we review the methods of constructing exact tests, the algorithm and commercial software. We also examine the difference between exact p-values obtained from exact tests and true p-values obtained from the true underlying distribution.