• Communications for Statistical Applications and Methods
• /
• v.10 no.3
• /
• pp.1057-1068
• /
• 2003

Results of using principal component analysis prior to cluster analysis are compared with results from applying agglomerative clustering algorithm alone. The retrieval ability of the agglomerative clustering algorithm is improved by using principal components prior to cluster analysis in some situations. On the other hand, the loss in retrieval ability for the agglomerative clustering algorithms decreases, as the number of informative variables increases, where the informative variables are the variables that have distinct information(or, necessary information) compared to other variables.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1996.10a
• /
• pp.147-150
• /
• 1996

A homing torpedo's performance can be expressed a function of many variables, i.e. technical and tactical variables. When designing a homing torpedo, these variables have to be decided upon. The system effectiveness of a homing torpedo can be determined by analyzing of these variables. This paper describes a procedure of simulation metamodelling using a Factor Analysis methodology. A simulation model was used in order to obtain the data base for analyzing detection probability of torpedo. By analyzing the main and interaction effects these variables on the analysis of detection probability, we will show the importance of certain variables, of a homing torpedo.

• Communications of the Korean Mathematical Society
• /
• v.13 no.4
• /
• pp.855-863
• /
• 1998

Some conditions on the strong law of large numbers for weighted sums of negative quadrant dependent random variables are studied. The almost sure convergence of weighted sums of negatively associated random variables is also established, and then it is utilized to obtain strong laws of large numbers for weighted averages of negatively associated random variables.

• Journal of applied mathematics & informatics
• /
• v.37 no.3_4
• /
• pp.207-217
• /
• 2019

We are presented of several basic properties for negatively superadditive dependent(NSD) random variables. By using this concept we are obtained complete convergence for maximum partial sums of rowwise NSD random variables. These results obtained in this paper generalize a corresponding ones for independent random variables and negatively associated random variables.

• Journal of the Ergonomics Society of Korea
• /
• v.22 no.4
• /
• pp.59-78
• /
• 2003

Most regression models of anthropometric variables use stature and/or weight as regressors; however, these 'flat' regression models result in large errors for anthropometric variables having low correlations with the regressors. To develop more accurate regression models for anthropometric variables, this study proposed a method to estimate anthropometric variables in a hierarchical manner based on the relationships among the variables and a process to develop and improve corresponding regression models. By applying the proposed approach, a hierarchical estimation structure was constructed for 59 anthropometric variables selected for the occupant package design of a passenger car and corresponding regression models were developed with the 1988 US Army anthropometric survey data. The hierarchical regression models were compared with the corresponding flat regression models in terms of accuracy. As results, the standard errors of the hierarchical regression models decreased by 28% (4.3mm) on average compared with those of the flat models.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.19 no.3
• /
• pp.434-447
• /
• 1995

This research was intended to identify variables influencing on risk perception and risk reduction behavior in clothing purchase situations. Responses from 631 female adults living in Seoul area were collected and analyzed. Towner for social occasions or working in office was used as clothing stimulus. The analysis included three product variables(price, style, type of clothing), 4 personality variables(generalized self-confidence, specific self-confidence, generalized informativeness, fashion informativeness), 2 clothing attitude variables(clothing importance, clothing interest), 4 demographic variables(age, educational level, occupation, income), and 3 situational variables(purchase planning, time pressure, effects of shopping company). Multiple regression revealed the fact that each type of clothing risk and each type of risk reduction behavior was influenced by the set of different variables. Generalized self- confidence and age and time pressure had more effects on clothing risk perception, while clothing risk reduction behavior was more influenced by clothing risk type, clothing interest, price of clothing and fashion informativeness. Implications for marketing strategies planning were also provided.

• Journal of Family Resource Management and Policy Review
• /
• v.5 no.1
• /
• pp.111-122
• /
• 2001

The propose of this study is to recognize the present consuming trends of adolescents and to verify the most influential variables affecting on the conspicuous consumption of adolescents. The samples were selected from the first and second grade level among male and female high school students in Seoul and Inchon area. The 384 subjects were analyzed by the SPSS/WIN program. The major findings of this study is as follows. First, the general tendency for conspicuous consumption of adolescents showed a lower level of scores than the average. Second, The conspicuous consumption of adolescents showed a significant differences, according to the socio-demographic variables, the psychological variables, the educational level of the adolescents, and the environmental variables. Third, the most influential variables affecting on the conspicuous consumption of adolescents are the variables, such as friends, mass media, sex, mother’s educational level, and material-oriented tendency. The total explonation of these variables was 54.7%.

• The Mathematical Education
• /
• v.43 no.2
• /
• pp.187-197
• /
• 2004

The present study investigated the effects of school and student background variables on academic achievement through analyzing Korean data from the OECD's Programme for International Student Assessment(PISA) which had been administered in 2000. The regression analysis was used to analyze the data, responses of 2,769 students(1,545 boys and 1,221 girls). Results of regression analysis were that 5.5% of process variables was accounted for by student variables and 0.3% of process variables was accounted for by school variables. So we try to find many other process variables which effect math achievements.

• Journal of Families and Better Life
• /
• v.17 no.3
• /
• pp.17-32
• /
• 1999

This study shows how the lifestyle of Housewife as well as usual socio-economic variables affect the expenditure of urban household. The Result of Factor Analysis revealed that the lifestyle of urban housewives were composed by four different traits. 1) self-improvement and rational trait 2) consumption-oriented convenience seeking trait 3) traditional-conservation trait 4) money-oriented variety seeking trait. Based on the Multiple regression analysis the analysis of the relative influence of variables which were associated with the expenditure of each item showed that influential variables for each item were different but economic variables(e, g , income saving, property etc) affected on most of expenditure. Socio-economic variables were more influential than lifestyle variables for every item. But lifestyle variables were influential the same as Socio-economic variables for unnecessary expenditure item more than necessary items particularly consumption-oriented convenience eeking trait and money-oriented variety seeking trait affected.

• Journal of the Korean Home Economics Association
• /
• v.47 no.5
• /
• pp.1-11
• /
• 2009

The purpose of this study was to examine the relationship between preschoolers’ self-perception, peer competence and giftedness. The subjects in this study were 101 preschoolers between the ages of 4 and 5 who attended childcare centers and private kindergartens in Pohang city area. The major findings of this study were as follows: Firstly, there were significant negative correlation between cognitive ability variables of self-perception and aggressive and asocial variables of peer competence. Secondly, there were significant positive correlation between cognitive ability variables of self-perception and humor variables of giftedness. Furthermore there were significant positive correlation between mother acceptance variables of selfperception and logical thinking, task commitment and humor variables of giftedness. Finally, there were significant negative correlation between hyper-distractible variables of peer competence and independence variables of giftedness.