• Communications for Statistical Applications and Methods
• /
• v.10 no.2
• /
• pp.507-517
• /
• 2003

Self-organizing map (SOM) has been developed mainly by T. Kohonen and his colleagues as a unsupervised learning neural network. Because of its topological ordering property, SOM is known to be very useful in pattern recognition and text information retrieval areas. Recently, data miners use Kohonen´s mapping method frequently in exploratory analyses of large data sets. One problem facing SOM builder is that there exists no sensible criterion for evaluating goodness-of-fit of the map at hand. In this short communication, we propose valid evaluation procedures for the Kohonen SOM of any size. The methods can be used in selecting the best map among several candidates.

• Journal of Korean Society for Quality Management
• /
• v.29 no.3
• /
• pp.49-65
• /
• 2001

This paper presents new compromise ALT plan which is applied to situations that true relationship between stress and parameters is not known exactly. The assumed failure distribution of this study is one of location-scale family, i. e., exponential, Weibull, and lognormal distributions which have been ones of the popular choices of failure distributions. The method of applying the stress is constant, and the censoring mechanism is Type I censoring. Compared with existing compromise plans under true simple linear model in terms of statistical efficiency, the efficiency of new compromise plan is better than the corresponding other compromise ones in most cases. For case when true model is quadratic, this plan can be used without any severe loss in statistical efficiency. The proposed new compromise ALT plan is illustrated with a numerical example and sensitivity analyses are conducted to study effects of pre-estimates of design parameters.

• Journal of the Korean Ceramic Society
• /
• v.34 no.9
• /
• pp.927-934
• /
• 1997

Static and cyclic fatigue tests were carried out for alumina specimen to study the statistical analyses (normal, lognormal and Weibull distribution) of fatigue lifetime data and nominal initial crack length data. Fatigue lifetime data followed Weibull distribution better than normal or lognormal distribution, for the shape parameter of the notched specimen was larger than that of the unnotched specimen. The nominal initial crack length data obtained from fatigue lifetime followed the lognormal and Weibull distribution better than normal distribution, for the coefficient of variation of the unnotched specimen was larger than that of the notched specimen, and shape parameter of unnotched specimen was smaller than that of the notched specimen.

• Journal of the Korean Home Economics Association
• /
• v.35 no.1
• /
• pp.415-428
• /
• 1997

In the present study we examined clothing expenditure patterns and related variables in Korea. In addition we analyzed the differences of clothing expenditure patterns between overall overspenders and non-overspenders. We use the Family Income and Expenditure Study published by the National Statistical Office of Korea. Double logarithm functional forms were used to adjust the normality of sample distribution and multiple regression analyses and t-test were utilized as a statistical tools. The present study was divided into four folders. First the income elasticity of clothing expenditures was examined by different groups such as age job and education levels of households as percentage change of clothing expenditures to a percentage change of income. Second to analyze the effects of demogtraphic and socio-economic variables on clothing expenditure we utlized the standardized coefficients in the separate regression equation by demographic and socio-economic variables. Third using spending to income ratio we investigated the differnces of the clothing expenditure patterns between overall overspenders and non-overspenders. Fourth the effects of independent variables by ovespenders and non-overspenders werediscussed.

• The Korean Journal of Applied Statistics
• /
• v.24 no.6
• /
• pp.1103-1113
• /
• 2011

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.

• Communications for Statistical Applications and Methods
• /
• v.26 no.6
• /
• pp.557-573
• /
• 2019

Mortality study is an essential component of actuarial risk management for life insurance policies, annuities, and pension plans. Life expectancy has drastically increased over the last several decades; consequently, longevity risk associated with annuity products and pension systems has emerged as a crucial issue. Among the various aspects of mortality study, a consideration of the cause-of-death mortality can provide a more comprehensive understanding of the nature of mortality/longevity risk. In this case study, the cause-of-mortality data in Korea and the US were analyzed along with a multinomial logistic regression model that was constructed to quantify the impact of mortality reduction in a specific cause on actuarial values. The results of analyses imply that mortality improvement due to a specific cause should be carefully monitored and reflected in mortality/longevity risk management. It was also confirmed that multinomial logistic regression model is a useful tool for analyzing cause-of-death mortality for actuarial applications.

• Transactions of Materials Processing
• /
• v.20 no.2
• /
• pp.124-131
• /
• 2011

A flexible stretch forming process is useful for small quantity batch production because various shape changes of the flexible die can be achieved conveniently. In this study, the design variables, namely, the punch size, curvature radius and elastic pad thickness, were quantitatively evaluated to understand their influence on sheet formability using statistical methods such as the correlation and regression analyses. Forming simulations were designed and conducted by a three-way factorial design to obtain numerical values of a shape error. Linear relationships between the design variables and the shape error resulted from the Pearson correlation analysis. Subsequently, a regression analysis was also conducted between the design variables and the shape error. A regression equation was derived and used in the flexible die design stage to estimate the shape error.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.12 no.4
• /
• pp.93-99
• /
• 2013

Parts and structures such as piston rod and knuckle joint for the use of hydraulic cylinder are often welded together in some fashion, usually due to cost and process effectiveness. Welding strongly affects the material by the process of heating and subsequent cooling as well as by the fusion process with additional filler material. Furthermore, a weld is usually far from being perfect, containing inclusions, pores, cavities, undercuts etc. As a consequence, fatigue failures appear in welded structures mostly at the welds rather than in the base metal, even if the latter contains notches. For this reason, fatigue analyses are of high practical interest for all welded structures under the action of cyclic loads. This paper describes the influence of welding parameters, material combinations and heat treatment on the fatigue behavior of welded cylinder rod. In addition, statistical characterization of stress-life response in weldment of hydraulic cylinder rod are presented.

• Journal of Information Technology Services
• /
• v.10 no.2
• /
• pp.203-221
• /
• 2011

The purpose of this study is to investigate the applications of Structural Equation Modeling(SEM) into MIS area in recent years. Two methodologies, Lisrel and PLS, are adopted for the method comparison. A research model, based upon TAM(Technology Acceptance Model) is used for the analysis of the data set of a previous study. The research model includes six research variables that are composed of twenty-eight question items. 272 data are used for data analyses through Lisrel v.8.72 and Visual PLS v.1.04. This study shows the statistical results of Lisrel are the same to those of PLS. The contribution of this study can be suggested as the followings; (1) A theoretical comparison of two methodologies is shown, (2) A statistical analysis is done at a real-situated data set, and (3) Several implications are suggested.

• Communications for Statistical Applications and Methods
• /
• v.12 no.1
• /
• pp.241-252
• /
• 2005

In this study we suggest that the spatial correlation structure of the brain fMRI data be used to characterize the functional connectivity of the brain. For some concussion and recovery data, we examine how the correlation structure changes from one step to another in the data analyses, which will allow us to see the effect of each analysis to the spatial correlation or the functional connectivity of the brain. This will lead us to spot the processes which cause significant changes in the spatial correlation structure of the brain. We discuss whether or not we can decompose correlation matrices in terms of its causes of variations in the data.