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 24, Issue 6 - Dec 2011
Volume 24, Issue 5 - Oct 2011
Volume 24, Issue 4 - Aug 2011
Volume 24, Issue 3 - Jun 2011
Volume 24, Issue 2 - Apr 2011
Volume 24, Issue 1 - Feb 2011
Selecting the target year
ROC Function Estimation
Hong, Chong-Sun ; Lin, Mei Hua ; Hong, Sun-Woo ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 987~994
DOI : 10.5351/KJAS.2011.24.6.987
From the point view of credit evaluation whose population is divided into the default and non-default state, two methods are considered to estimate conditional distribution functions: one is to estimate under the assumption that the data is followed the mixture normal distribution and the other is to use the kernel density estimation. The parameters of normal mixture are estimated using the EM algorithm. For the kernel density estimation, five kinds of well known kernel functions and four kinds of the bandwidths are explored. In addition, the corresponding ROC functions are obtained based on the estimated distribution functions. The goodness-of-fit of the estimated distribution functions are discussed and the performance of the ROC functions are compared. In this work, it is found that the kernel distribution functions shows better fit, and the ROC function obtained under the assumption of normal mixture shows better performance.
A Hilbert-Huang Transform Approach Combined with PCA for Predicting a Time Series
Park, Min-Jeong ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 995~1006
DOI : 10.5351/KJAS.2011.24.6.995
A time series can be decomposed into simple components with a multiscale method. Empirical mode decomposition(EMD) is a recently invented multiscale method in Huang et al. (1998). It is natural to apply a classical prediction method such a vector autoregressive(AR) model to the obtained simple components instead of the original time series; in addition, a prediction procedure combining a classical prediction model to EMD and Hilbert spectrum is proposed in Kim et al. (2008). In this paper, we suggest to adopt principal component analysis(PCA) to the prediction procedure that enables the efficient selection of input variables among obtained components by EMD. We discuss the utility of adopting PCA in the prediction procedure based on EMD and Hilbert spectrum and analyze the daily worm account data by the proposed PCA adopted prediction method.
Comparison Study on the Performances of NLL and GMM for Estimating Diffusion Processes
Kim, Dae-Gyun ; Lee, Yoon-Dong ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1007~1020
DOI : 10.5351/KJAS.2011.24.6.1007
Since the research of Black and Scholes (1973), modeling methods using diffusion processes have performed principal roles in financial engineering. In modern financial theories, various types of diffusion processes were suggested and applied in real situations. An estimation of the model parameters is an indispensible step to analyze financial data using diffusion process models. Many estimation methods were suggested and their properties were investigated. This paper reviews the statistical properties of the, Euler approximation method, New Local Linearization(NLL) method, and Generalized Methods of Moment(GMM) that are known as the most practical methods. From the simulation study, we found the NLL and Euler methods performed better than GMM. GMM is frequently used to estimate the parameters because of its simplicity; however this paper shows the performance of GMM is poorer than the Euler approximation method or the NLL method that are even simpler than GMM. This paper shows the performance of the GMM is extremely poor especially when the parameters in diffusion coefficient are to be estimated.
A Case Study on the Risk of Stepdown ELS
Kim, Hee-Sun ; Yeo, In-Kwon ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1021~1031
DOI : 10.5351/KJAS.2011.24.6.1021
Equity linked securities are indirect investments where the return of investment depends on the performance of the underlying equities. In this paper, we review the profit structure of typical equity linked securities through a profit diagram and investigate which characteristics of time series at the investment affect the early repayment of the stepdown ELS based on KOSPI 200 and HSI. We also compare VaRs using the empirical distribution function for risk management.
Empirical Analyses of Asymmetric Conditional Heteroscedasticities for the KOSPI and Korean Won-US Dollar Exchange Rate
Maeng, Hye-Young ; Shin, Dong-Wan ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1033~1043
DOI : 10.5351/KJAS.2011.24.6.1033
In this paper, we use a nested family of models of Generalized Autoregressive Conditional Heteroscedasticity(GARCH) to verify asymmetric conditional heteroscedasticity in the KOSPI and Won-Dollar exchange rate. This study starts from an investigation of whether time series data have asymmetric features not explained by standard GARCH models. First, we use kernel density plot to show the non-normality and asymmetry in data as well as to capture asymmetric conditional heteroscedasticity. Later, we use three representative asymmetric heteroscedastic models, EGARCH(Exponential Garch), GJR-GARCH(Glosten, Jagannathan and Runkle), APARCH(Asymmetric Power Arch) that are improved from standard GARCH models to give a better explanation of asymmetry. Thereby we highlight the fact that volatility tends to respond asymmetrically according to positive and/or negative values of past changes referred to as the leverage effect. Furthermore, it is verified that how the direction of asymmetry is different depending on characteristics of time series data. For the KOSPI and Korean won-US dollar exchange rate, asymmetric heteroscedastic model analysis successfully reveal the leverage effect. We obtained predictive values of conditional volatility and its prediction standard errors by using moving block bootstrap.
A Composite Trend Test with Symptom Occurrence and Severity Symptom Scores
Choi, Se-Mi ; Yang, Soo ; Song, Hae-Hiang ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1045~1054
DOI : 10.5351/KJAS.2011.24.6.1045
During clinical trials a researcher is frequently able to observe a disease symptom in a subject as well as a severity score for those who experienced a symptom after a fixed length of treatment. The traditional method to evaluate a decreasing trend in proportion, when there is an intrinsic order in the treatment groups (for example control and two or more treatment groups) is a Cochran-Armitage test, while the method to evaluate a decreasing trend in continuous non-normal data is a Jonckheere-Tersptra test. The Cochran-Armitage test emphasizes the dichotomous data of symptom occurrence and the Jonckheere-Tersptra test emphasizes the continuous non-normal data of severity symptom scores. In this paper we propose new test statistics that consider the combined evidence from a symptom occurrence and disease severity score. We illustrate these methods with example data of schizophrenic inpatients that demonstrated antipsychotic-drug induced constipation. A small-scale simulation is conducted to compare the new trend tests with other trend tests.
On Evaluation of Bioequivalence for Highly Variable Drugs
Jeong, Gyu-Jin ; Park, Sang-Gue ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1055~1076
DOI : 10.5351/KJAS.2011.24.6.1055
This paper reviews the definition of highly variable drug(HVD), the present regulatory recommendations and the approaches proposed in the literature to deal with the bioequivalence issues of HVD. The concept and the statistical approach of scaled average bioequivalence(SABE) is introduced and discussed with the current regulatory methods. The recommendations for SABE approach are proposed and the further study topics related to HVDs are also presented.
Gene Screening and Clustering of Yeast Microarray Gene Expression Data
Lee, Kyung-A ; Kim, Tae-Houn ; Kim, Jae-Hee ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1077~1094
DOI : 10.5351/KJAS.2011.24.6.1077
We accomplish clustering analyses for yeast cell cycle microarray expression data. To reflect the characteristics of a time-course data, we screen the genes using the test statistics with Fourier coefficients applying a FDR procedure. We compare the results done by model-based clustering, K-means, PAM, SOM, hierarchical Ward method and Fuzzy method with the yeast data. As the validity measure for clustering results, connectivity, Dunn index and silhouette values are computed and compared. A biological interpretation with GO analysis is also included.
Bayesian Estimation in Bioequivalence Study
Lee, Seung-Chun ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1095~1102
DOI : 10.5351/KJAS.2011.24.6.1095
The classical two-period, two-sequence crossover design is no longer sufficient to assess various demands in a bioequivalence study. For instance, to estimate the within-subject and between-subject variances of test and reference formulations separately, it is necessary to use a replicate design in which each subject receives at least the reference formulation in two periods. Several designs were studied to satisfy the demands. It is provided a unified Bayesian approach applicable to those study designs. The benefit of the method in the bioequivalence study is discussed.
Performance Comparison of Classication Methods with the Combinations of the Imputation and Gene Selection Methods
Kim, Dong-Uk ; Nam, Jin-Hyun ; Hong, Kyung-Ha ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1103~1113
DOI : 10.5351/KJAS.2011.24.6.1103
Gene expression data is obtained through many stages of an experiment and errors produced during the process may cause missing values. Due to the distinctness of the data so called `small n large p`, genes have to be selected for statistical analysis, like classification analysis. For this reason, imputation and gene selection are important in a microarray data analysis. In the literature, imputation, gene selection and classification analysis have been studied respectively. However, imputation, gene selection and classification analysis are sequential processing. For this aspect, we compare the performance of classification methods after imputation and gene selection methods are applied to microarray data. Numerical simulations are carried out to evaluate the classification methods that use various combinations of the imputation and gene selection methods.
A Composite Estimator for the Take-Nothing Stratum of Cut-Off Sampling
Kim, Ji-Hak ; Shin, Key-Il ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1115~1128
DOI : 10.5351/KJAS.2011.24.6.1115
Cut-off sampling that discards a part of the population from the sampling frame, is a widely used method for a highly skewed population like a business survey. Usually to the estimate of population total, we need to estimate the total of the take-nothing stratum. Many estimators have been developed to estimate the total of the take-nothing stratum. In this paper, we suggest a new composite estimator which combines the estimator suggested by Sarndal et al. (1992) and a ratio estimator obtained by small samples from the take-nothing stratum. Small simulation studies are performed for the comparison of the estimators and we confirm that the new suggested estimator is superior to the others.
An Efficient Estimation of Local Area Unemployment Rate Based on Small Area Estimation
Kim, Soo-Taek ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1129~1138
DOI : 10.5351/KJAS.2011.24.6.1129
Small area estimation has received significant intention in recent years due to a growing demand for reliable local area statistics. Traditional area-specific direct estimates based solely on sample survey data in the areas of interest do not provide adequate small area precision; however, design-based indirect local area estimators borrow strength from sample observations of related areas to increase the effective sample size. Design-based indirect estimation methods such as synthetic and composite estimators are considered to adjust local area unemployment rate estimates in the Korean Economically Active Population Survey. This study suggests an efficient alternative to minimize the cost to construct the unemployment rate of a local area through simulation under the condition that we can maintain a certain level of CV for the estimates. We obtained the results that the composite estimators using a sample size greater than 10 are more stable and significant at the level of CV 25% in our design scheme.
Suggestion of a New Method of Computing Percentage of Victories for the Korean Professional Baseball
Kim, Hyuk-Joo ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1139~1148
DOI : 10.5351/KJAS.2011.24.6.1139
Team standings in the regular professional baseball league should be determined based on a reasonable criterion; however, an unreasonable Japanese method is being used in Korea as of 2011. In this paper, we suggest a new method of computing the percentage of victories constructed by combining the advantages of the methods to determine team standings used in Korean professional baseball. We also have applied preexistent methods and suggested method to past and present Korean professional baseball data.
Inferential Problems in Bayesian Logistic Regression Models
Hwang, Jin-Soo ; Kang, Sung-Chan ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1149~1160
DOI : 10.5351/KJAS.2011.24.6.1149
Model selection and hypothesis testing problems in Bayesian inference are still debated between scholars. Bayesian factors traditionally used as a criterion in Bayesian hypothesis testing and model selection, are easy to understand but sometimes hard to compute. In addition, there are other model selection criterions such as DIC(Deviance Information Criterion) by Spiegelhalter et al. (2002) and Bayesian P-values for testing. In this paper, we briefly introduce the Bayesian hypothesis testing and model selection procedure. In addition we have applied a Bayesian inference to Swiss banknote data by a fitting logistic regression model and computing several test statistics to see if they provide consistent results.
Polynomially Adjusted Normal Approximation to the Null Distribution of Ansari-Bradley Statistic
Ha, Hyung-Tae ; Yang, Wan-Youn ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1161~1168
DOI : 10.5351/KJAS.2011.24.6.1161
The approximation for the distribution functions of nonparametric test statistics is a significant step in statistical inference. A rank sum test for dispersions proposed by Ansari and Bradley (1960), which is widely used to distinguish the variation between two populations, has been considered as one of the most popular nonparametric statistics. In this paper, the statistical tables for the distribution of the nonparametric Ansari-Bradley statistic is produced by use of polynomially adjusted normal approximation as a semi parametric density approximation technique. Polynomial adjustment can significantly improve approximation precision from normal approximation. The normal-polynomial density approximation for Ansari-Bradley statistic under finite sample sizes is utilized to provide the statistical table for various combination of its sample sizes. In order to find the optimal degree of polynomial adjustment of the proposed technique, the sum of squared probability mass function(PMF) difference between the exact distribution and its approximant is measured. It was observed that the approximation utilizing only two more moments of Ansari-Bradley statistic (in addition to the first two moments for normal approximation provide) more accurate approximations for various combinations of parameters. For instance, four degree polynomially adjusted normal approximant is about 117 times more accurate than normal approximation with respect to the sum of the squared PMF difference.
A Study on the Frequency Structure of Probability Distributions Using Social Network Analysis
Jang, Dae-Heung ; Yi, Seong-Baek ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1169~1179
DOI : 10.5351/KJAS.2011.24.6.1169
Through social network analysis using portal site information, we study the relation of the probability distributions that appear in statistics textbooks with probability distributions that appear in daily life. Based on daily life, we discuss probability distributions that must be emphasized in frequent use.
Robust Image Fusion Using Stationary Wavelet Transform
Kim, Hee-Hoon ; Kang, Seung-Hyo ; Park, Jea-Hyun ; Ha, Hyun-Ho ; Lim, Jin-Soo ; Lim, Dong-Hoon ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1181~1196
DOI : 10.5351/KJAS.2011.24.6.1181
Image fusion is the process of combining information from two or more source images of a scene into a single composite image with application to many fields, such as remote sensing, computer vision, robotics, medical imaging and defense. The most common wavelet-based fusion is discrete wavelet transform fusion in which the high frequency sub-bands and low frequency sub-bands are combined on activity measures of local windows such standard deviation and mean, respectively. However, discrete wavelet transform is not translation-invariant and it often yields block artifacts in a fused image. In this paper, we propose a robust image fusion based on the stationary wavelet transform to overcome the drawback of discrete wavelet transform. We use the activity measure of interquartile range as the robust estimator of variance in high frequency sub-bands and combine the low frequency sub-band based on the interquartile range information present in the high frequency sub-bands. We evaluate our proposed method quantitatively and qualitatively for image fusion, and compare it to some existing fusion methods. Experimental results indicate that the proposed method is more effective and can provide satisfactory fusion results.
Comparison of Structural Change Tests in Linear Regression Models
Kim, Jae-Hee ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1197~1211
DOI : 10.5351/KJAS.2011.24.6.1197
The actual power performance of historical structural change tests are compared under various alternatives. The tests of interest are F, CUSUM, MOSUM, Moving Estimates and empirical distribution function tests with both recursive and ordinary least-squares residuals. Our comparison of the structural tests involves limiting distributions under the hypothesis, the ability to detect the alternative hypotheses under one or double structural change, and smooth change in parameters. Even though no version is uniformly superior to the other, the knowledge about the properties of those tests and connections between these tests can be used in practical structural change tests and in further research on other change tests.
Variable Selection in Normal Mixture Model Based Clustering under Heteroscedasticity
Kim, Seung-Gu ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1213~1224
DOI : 10.5351/KJAS.2011.24.6.1213
In high dimensionality where the number of variables are excessively larger than observations, it is required to remove the noninformative variables to cluster observations. Most model-based approaches for variable selection have been considered under the assumption of homoscedasticity and their models are mainly estimated by a penalized likelihood method. In this paper, a different approach is proposed to remove the noninformative variables effectively and to cluster based on the modified normal mixture model simultaneously. The validity of the model was provided and an EM algorithm was derived to estimate the parameters. Simulation studies and an experiment using real microarray dataset showed the effectiveness of the proposed method.
A Statistical Tuning Method to Improve the Accuracy of 1Km×1Km Resolution-Wind Data of South Korea Generated from a Numerical Meteorological Model
Kim, Hea-Jung ; Kim, Hyun-Sik ; Choi, Young-Jean ; Lee, Seong-Woo ; Seo, Beom-Keun ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1225~1235
DOI : 10.5351/KJAS.2011.24.6.1225
This paper suggests a method for tuning a numerically simulated wind speed data, provided by NIMR(National Institute of Meteorological Research) and generated from a numerical meteorological model to improve a wind resource map with a
resolution. To this end, "tuning factor method" is developed that consists of two procedures. First, estimate monthly wind fields based on a suitably designed statistical wind field model that covers 345,682 regions obtained by
lattice sites in South Korea. The second procedure computes the tuning factor and then tunes the generated wind speeds of each month as well as each lattice site. The second procedure is based on the wind fields estimated by the first procedure. The performance of the suggested tuning method is demonstrated by using two wind data(both TMY and numerically simulated wind speed data) of 75 weather station areas.
Analysis and Application to Customers` Social Roles Using Voice Network of a Telecom Company
Chun, Heui-Ju ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1237~1248
DOI : 10.5351/KJAS.2011.24.6.1237
Social network analysis(SNA) has been recently applied to business areas such as social network services (such as Facebook and Twitter). In addition, the mobile telecommunication field attempts to analyze CDR(call detail record) data and apply customer relationship management and customer churn management through the use of social network analysis. The paper analyzes links between ego and alter based on ego-network and discovers four kinds of customer roles and then provides insights as a tool for customer relationship management or customer management.
On the Use of Sequential Adaptive Nearest Neighbors for Missing Value Imputation
Park, So-Hyun ; Bang, Sung-Wan ; Jhun, Myoung-Shic ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1249~1257
DOI : 10.5351/KJAS.2011.24.6.1249
In this paper, we propose a Sequential Adaptive Nearest Neighbor(SANN) imputation method that combines the Adaptive Nearest Neighbor(ANN) method and the Sequential k-Nearest Neighbor(SKNN) method. When choosing the nearest neighbors of missing observations, the proposed SANN method takes the local feature of the missing observations into account as well as reutilizes the imputed observations in a sequential manner. By using a Monte Carlo study and a real data example, we demonstrate the characteristics of the SANN method and its potential performance.
Analysis of Papers in the Korean Journal of Applied Statistics by Co-Author Networks Analysis
Lee, M. ; Park, M. ; Lee, H. ; Jin, S. ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1259~1270
DOI : 10.5351/KJAS.2011.24.6.1259
This study analyzed an aspect of co-author relationship in papers published in the Korean Journal of Applied Statistics by social network analysis. The data were extracted from 664 papers in the journal from 2000 to 2010. Authors at center of the network are detected by a network centrality analysis. Sub-network analysis found distinguishable research groups from the point of view of their topics or affiliations. The significance of affiliations to co-author relationship was examined by logistic regression analysis.
Kullback-Leibler Information-Based Tests of Fit for Inverse Gaussian Distribution
Choi, Byung-Jin ;
Korean Journal of Applied Statistics, volume 24, issue 6, 2011, Pages 1271~1284
DOI : 10.5351/KJAS.2011.24.6.1271
The entropy-based test of fit for the inverse Gaussian distribution presented by Mudholkar and Tian(2002) can only be applied to the composite hypothesis that a sample is drawn from an inverse Gaussian distribution with both the location and scale parameters unknown. In application, however, a researcher may want a test of fit either for an inverse Gaussian distribution with one parameter known or for an inverse Gaussian distribution with both the two partameters known. In this paper, we introduce tests of fit for the inverse Gaussian distribution based on the Kullback-Leibler information as an extension of the entropy-based test. A window size should be chosen to implement the proposed tests. By means of Monte Carlo simulations, window sizes are determined for a wide range of sample sizes and the corresponding critical values of the test statistics are estimated. The results of power analysis for various alternatives report that the Kullback-Leibler information-based goodness-of-fit tests have good power.