• Journal of the Korean Statistical Society
• /
• v.35 no.2
• /
• pp.213-224
• /
• 2006

We investigate the influence of subjects or observations on regression coefficients of generalized estimating equations using the influence function and the derivative influence measures. The influence function for regression coefficients is derived and its sample versions are used for influence analysis. The derivative influence measures under certain perturbation schemes are derived. It can be seen that the influence function method and the derivative influence measures yield the same influence information. An illustrative example in longitudinal data analysis is given and we compare the results provided by the influence function method and the derivative influence measures.

• Journal of Information Science Theory and Practice
• /
• v.1 no.2
• /
• pp.36-57
• /
• 2013

Blogs are readily available sources of opinions and sentiments which allows bloggers to exert a certain level of influence over the blog readers. Previous studies had attempted to analyze blog features to detect influence within the blogosphere, but had not studied in details influence at the blogger-level. Other studies studied bloggers' personalities with regards to their propensity to blog, but did not relate the personalities of bloggers to influence. Bloggers may differ in their way or manner of exerting influence. For example, bloggers could be active participants or just passive shares, or whether they express ideas in a rational or subjective manner, or they are received positively or negatively by the readers. In this paper, we further analyze the engagement style (frequency, scope, originality, and consistency of the blog postings), persuasion style (appeals to reasons or emotions), and persona (degree of compliance) of individual bloggers. Methods used include similarity analysis to detect the sharing-creating aspect of engagement style, subjectivity analysis to measure persuasion style, and sentiment analysis to identify persona style. While previous studies analyzed influence at blog site level, our model is shown to provide a fine-grained influence analysis that could further differentiate the bloggers' influence style in a blog site.

• Journal of the Korean Statistical Society
• /
• v.27 no.4
• /
• pp.471-483
• /
• 1998

The influence of observations on the misclassification probability in multiple discriminant analysis under the equal covariance assumption is investigated by the local influence method. Under an appropriate perturbation we can get information about influential observations and outliers by studying the curvatures and the associated direction vectors of the perturbation-formed surface of the misclassification probability. We show that the influence function method gives essentially the same information as the direction vector of the maximum slope. An illustrative example is given for the effectiveness of the local influence method.

• Journal of Power System Engineering
• /
• v.21 no.5
• /
• pp.48-54
• /
• 2017

The transfer influence coefficient method which is an vibration analysis algorithm based on the transfer of influence coefficient is applied to the free vibration analysis of double cylindrical shells. After the computational programs for the free vibration analysis of double cylindrical shells were made using the transfer influence coefficient method and the transfer matrix method, we compared the results using the transfer influence coefficient method with those by the transfer matrix method. The transfer influence coefficient method provided the good computational results in the free vibration analysis of double cylindrical shells. In particular, The results of the transfer influence coefficient method are superior to those of the transfer matrix method when the stiffness of internal springs connecting a inside cylindrical shell and a outside cylindrical shell is very large.

• Fashion & Textile Research Journal
• /
• v.15 no.6
• /
• pp.923-931
• /
• 2013

The purpose of this study was to examine the effect of the type of masculinity on appearance-oriented and appearance management behavior. Questionnaires were administered to 201 males 20s to 50s living in Daegu. The SPSS 20.0 package was utilized for data analysis which included frequency analysis, Cronbach's ${\alpha}$, factor analysis, cluster analysis, regression analysis, t-test and logistic regression. First, factor analysis in a appearance management behavior and the type of masculinity, were cluster analysis in appearance-oriented. Were used to analyze the problem of the study by classified five factors and two groups. The results of this study were as follows: First, the masculinity of Retro sexual, Metro sexual, Techno sexual showed a negative influence on appearance-oriented. Second, the masculinity had a influence on appearance management behavior. M-ness showed a negative influence on skin care, Metro sexual showed a positive influence on skin care, hair/fashion, cosmetics, plastic surgery. And Retro sexual showed a influence on skin care, positive influence on hair/fashion, cosmetics. Techno sexual showed a positive influence on body management, skin care, hair/fashion. Uber sexual showed a positive influence on body management, cosmetics of appearance management. Third, the appearance-oriented had a differences on body management, hair/fashion, cosmetics, plastic surgery of appearance management behavior.

• Journal of applied mathematics & informatics
• /
• v.5 no.2
• /
• pp.489-498
• /
• 1998

The influence of observations the on the goodness-of-fit test in maximum likelihood factor analysis is investigated by using the local influence method. under an appropriate perturbation the test statistic forms a surface. One of main diagnostics is the maximum slope of the perturbed surface the other is the direction vector cor-responding to the curvature. These influence measures provide the information about jointly influence measures provide the information about jointly influential observations as well as individ-ually influential observations.

• Journal of the Korean Statistical Society
• /
• v.30 no.3
• /
• pp.499-509
• /
• 2001

We investigate the influence of observations on a test of additional information about discrimination using the influence function and the derivative influence measures. the influence function for the test statistic is derived and this sample versions are used for influence analysis. The derivative influence measures for the test statistic under a perturbation scheme are derived. It will be seen that the influence function method and the derivative influence measures yield the same result. Furthermore, we will derive the relationships between the influence function and the derivative influence measures when the sample size is large. an illustrative example is given and we will compare the results provided by the influence function method and the derivative influence measures.

• Korean journal of food and cookery science
• /
• v.29 no.5
• /
• pp.533-541
• /
• 2013

The purpose of this article was to study the influence of tangible clues of raw-fish restaurants on consumers' emotional responses and repurchasing behavior in eastern area of Kangwon-province. In this study frequency analysis, reliability analysis, factor analysis and path analysis(SEM) were applied to analyze the data. Through path analysis tangible clues' performances had generally significant influences upon consumers' emotional responses to buy products of raw-fish restaurants. According to the results, exterior facility had positive influence on pleasure(p<0.001) and arousal(p<0.01). Interior facility had positive influence on pleasure(p<0.001) and arousal(p<0.001), but negative on dominance(p<0.01). And the others had positive influence on pleasure(p<0.05). Press copy had positive influence on pleasure(p<0.001) and arousal(p<0.001). And Pleasure and arousal had positive influence on repurchasing behavior(p<0.001). But Dominance didn't have significant influence on repurchasing behavior. As a result, we studied that the tangible clues was an effective marketing tactic to maintain and increase consumers' positive emotional reponses to repurchase at restaurants. Therefore it is finally needed that dining-out corporations manage tangible clues in a customer-maintaining without additional costs, and especially to use mixing tangible clues each other in some suitable condition.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.38 no.6
• /
• pp.810-822
• /
• 2014

This study investigated the effects of sociocultural influence and body image on appearance enhancement behavior (facial management, clothing selection, and weight/figure management). For data collection, a questionnaire was administrated to 378 female college students in Seoul and Gyeonggi-do from May $23^{rd}$ to June $10^{th}$ 2013. A SPSS 18.0 statistics package was used to analyze data along with descriptive statistical analysis, frequency analysis, factor analysis, reliability analysis, and regression analysis and frequency analysis. The results were as follows. First, sociocultural influences were divided into three factors: media influence, peer influence, and parental influence. Overall sociocultural influences had positive effects on appearance enhancement behavior. Second, body image was divided into: appearances-management, body-satisfaction and body confidence. Sociocultural influences had a significant effect on overall body image. Third, body image positively affected overall appearance enhancement behavior.

• Journal of the Korea Institute of Building Construction
• /
• v.13 no.4
• /
• pp.424-433
• /
• 2013

There have been few attempts made to perform a systematic analysis of the various risk influence factors in building development projects. This study suggests an analysis of the risk influence factors in the process of feasibility studies for apartment building development projects. To reflect the voice of professionals, surveys were carried out. In addition, an FD-AHP method was applied to identify the importance of the risk influence factors. Through the surveys, major risk factors were separately identified as direct and/or indirect elements. An analysis of risk influence factors supports an effective feasibility study of apartment building development projects.