• Journal of the Korean Statistical Society
• /
• v.29 no.2
• /
• pp.201-217
• /
• 2000

The method of generalized estimating equations (GEE) has become very popular for the analysis of longitudinal data. We extend this work to the use of M-estimators; the resultant regression estimates are robust to heavy tailed errors and to outliers. The proposed method does not require correct specification of the dependence structure between observation, and allows for heterogeneity of the error. However, an estimate of the dependence structure may be incorporated, and if it is correct this guarantees a higher efficiency for the regression estimators. A goodness-of-fit test for checking the adequacy of the assumed M-estimation regression model is also provided. Simulation studies are conducted to show the finite-sample performance of the new methods. The proposed methods are applied to a real-life data set.

• Journal of the Korea Convergence Society
• /
• v.11 no.5
• /
• pp.97-105
• /
• 2020

This study was to identify the main predictors of smartphone overdependece of infants perceived by mothers and to provide basic data useful for education and practice. For this, data-mining decision tree analysis was performed using SPSS program, and the fianl 410 researches were used for analysis. The results. In the case of the whole infants, the most important predictor of smartphone dependence was father's leisure activity parenting participation. For boys, their father's leisure activity was the most dependent on their smartphone dependence. However, even if father's participation in leisure activities was high, smartphone overdependence increased again when mother's overprotection and permissive attitude were high. Finally, For girls, the most influential variable on smartphone dependence was warmth and encouragement among mothers' parenting attitudes.

• Tuberculosis and Respiratory Diseases
• /
• v.85 no.4
• /
• pp.349-357
• /
• 2022

Background: The most consistently identified mortality determinants for the new coronavirus 2019 (COVID-19) infection are aging, male sex, cardiovascular/respiratory diseases, and cancer. They were determined from heterogeneous cohorts that included patients with different disease severity and previous conditions. The main goal of this study was to determine if activities of daily living (ADL) dependence measured by Barthel's index could be a predictor for COVID-19 mortality. Methods: A prospective cohort study was performed with a consecutive sample of 340 COVID-19 patients representing patients from all over the northern region of Portugal from October 2020 to March 2021. Mortality risk factors were determined after controlling for demographics, ADL dependence, admission time, comorbidities, clinical manifestations, and delay-time for diagnosis. Central tendency measures were used to analyze continuous variables and absolute numbers (proportions) for categorical variables. For univariable analysis, we used t test, chi-square test, or Fisher exact test as appropriate (α=0.05). Multivariable analysis was performed using logistic regression. IBM SPSS version 27 statistical software was used for data analysis. Results: The cohort included 340 patients (55.3% females) with a mean age of 80.6±11.0 years. The mortality rate was 19.7%. Univariate analysis revealed that aging, ADL dependence, pneumonia, and dementia were associated with mortality and that dyslipidemia and obesity were associated with survival. In multivariable analysis, dyslipidemia (odds ratio [OR], 0.35; 95% confidence interval [CI], 0.17-0.71) was independently associated with survival. Age ≥86 years (pooled OR, 2.239; 95% CI, 1.100-4.559), pneumonia (pooled OR, 3.00; 95% CI, 1.362-6.606), and ADL dependence (pooled OR, 6.296; 95% CI, 1.795-22.088) were significantly related to mortality (receiver operating characteristic area under the curve, 82.1%; p<0.001). Conclusion: ADL dependence, aging, and pneumonia are three main predictors for COVID-19 mortality in an elderly population.

• Journal of the Korea Convergence Society
• /
• v.12 no.12
• /
• pp.583-594
• /
• 2021

This study is a descriptive research study conducted to understand the effects of hospital nurses' COVID-19 risk perception, media dependence, government trust, resilience, and Socio-psychological stress on COVID-19 prevention behavior. Data were collected from 200 nurses at university hospitals, and the collected data were analyzed using descriptive statistics, t-test, Pearson's correlation analysis, ANOVA, and multiple regression analysis using the SPSS 24.0 program. The average age of the subjects was 29.8 years old, the hospital experience was 6.6 years, and the prevention of COVID-19 was high with 4.3 points. COVID-19 prevention behavior showed positive correlation with COVID-19 risk perception, resilience, and media dependence and negative correlation with socio-psychological stress. Factors influencing COVID-19 prevention behavior were COVID-19 risk perception, resilience, socio-psychological stress, and media dependence. To ensure that hospital nurses' COVID-19 prevention behaviors can be continued, it is necessary to appropriately manage COVID-19 risk perception and media dependence, and to develop and apply an intervention program to strengthen resilience and reduce socio-psychological stress.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.4
• /
• pp.213-221
• /
• 2022

The purpose of this study is to find methods to prevent and improve smart device over-dependence problems by analyzing differences in school life adaptation and academic performance according to children's dependence on smart devices. For this, the data of fifth grade elementary school students in the 12th year were extracted and utilized from Panel Survey of Korean Children. The data were analyzed using non-hierarchical cluster(K-means) analysis, T-test, one-way ANOVA, and Scheffé tests. The results of this study are as follows. First, It has been shown that dependence on smart devices, school adaptation and academic performance have a negative correlation. Second, students in potential and high-risk groups who are highly dependent on smart devices have significantly lower school adaptation compared to those in the safety group. Third, high-risk students showed significantly lower academic performance compared to those in the potential risk group and general group. Based on these findings, it was suggested that for elementary school students who rely on smart devices, various learning support and national efforts such as counseling for school life adaptation are needed.

• Communications for Statistical Applications and Methods
• /
• v.28 no.1
• /
• pp.59-79
• /
• 2021

Value at Risk (VaR) is one of the most common risk management tools in finance. Since a portfolio of several assets, rather than one asset portfolio, is advantageous in the risk diversification for investment, VaR for a portfolio of two or more assets is often used. In such cases, multivariate distributions of asset returns are considered to calculate VaR of the corresponding portfolio. Copulas are one way of generating a multivariate distribution by identifying the dependence structure of asset returns while allowing many different marginal distributions. However, they are used mainly for bivariate distributions and are not widely used in modeling joint distributions for many variables in finance. In this study, we would like to examine the performance of various copulas for high dimensional data and several different dependence structures. This paper compares copulas such as elliptical, vine, and hierarchical copulas in computing the VaR of portfolios to find appropriate copula functions in various dependence structures among asset return distributions. In the simulation studies under various dependence structures and real data analysis, the hierarchical Clayton copula shows the best performance in the VaR calculation using four assets. For marginal distributions of single asset returns, normal inverse Gaussian distribution was used to model asset return distributions, which are generally high-peaked and heavy-tailed.

• The Journal of Industrial Distribution & Business
• /
• v.9 no.3
• /
• pp.45-56
• /
• 2018

Purpose - The Korean university education system is facing innovation and change, including cooperation between industry and university, Therefore It is important to activate the industry-university cooperation. This paper aims to demonstrate the factors that activate industry-university cooperation, particularly about the voluntary participation induction by industry and researching in path dependency perspectives. Research design, data, and methodology - The subject of this research were companies that are aware of the industry-university cooperation program. This research hypothesis is derived from the literature of previous studies of industry-university cooperation, This study have constructs that was defined operationally with reference to previous studies, this research model design to figure out structural relationship among technology leadership of university, university specialization, local network strength, fixation of local economy, recognition of path dependence, participation by industry, performance of industry-university cooperation. From 2017 July. 1 to Sept. 31, questionnaire survey targeting company staff who is involving in industry-university cooperation. 257 questionnaire survey had conducted. 249 investigated data were used for empirical analysis except wrong data. This data were used for AMOS(structural equation) & Regression statistics to verify hypothesis which developed by researcher. Results - The results of this study are as follows. First, technology leadership of universities has a significant effect on voluntary participation by industry. University specialization has significant effect on voluntary participation by industry. Second, local network strength has significant effect on voluntary participation by industry. but fixation of local economy does not affect voluntary participation by industry. Third, recognition of path dependence has moderating effect between Independent(university, company characteristics) and dependent variables(voluntary participation by industry) When recognition level of path dependence is high, preceding factors have a significant effect on voluntary participation by industry than recognition level of path dependence is low. As a result, the degree of recognition of path dependence was shown important variables that induce voluntary participation of industry for industry-university cooperation program. Conclusions - This study suggests that voluntary participation of industry is a very important factor in the achievement of industry-university cooperation. Recognition of interdependence as well as leading factors that encourage voluntary participation of industry is also just as important. If recognition of path dependence was high, Industry's voluntary participation was high.

• Journal of the Korean Institute of Gas
• /
• v.1 no.1
• /
• pp.7-13
• /
• 1997

To investigate the pressure dependence of minimum ignition energy(MIE), thermal ignition theory, concept of heat transfer, ideal gas law, and kinetic theory are discussed. Correlation equations for the MIE and pressure were obtained through a regression analysis of reported data. In the proposed methodology the predicted MIE with pressure variations agree with reported data within a few average absolute deviations(A.A.D.). Therefore the proposed methodology has provided to be the general method for predicting the MIE of hydrocarbons.

• Asian Pacific Journal of Cancer Prevention
• /
• v.13 no.4
• /
• pp.1285-1288
• /
• 2012

Background: The Fagerstrom Test for Cigarette Dependence (FTCD) (formally FTND) is widely used for measuring physical dependence on nicotine. Objective: To explore the cross cultural validity and reliability of FTCD amongst Arabic speaker cigarette consumers who chew khat leaf, a stimulant green leaf. Methods: The psychometric properties of the FTCD were assessed in a subsample (91regular cigarette smokers) of purposively selected 204 UK-resident Yemeni khat chewers recruited during random visits to khat sale outlets. Data were collected via a structured face-to-face interview. Data analyses included descriptive tests and factor analysis. Results: Two factors were obtained by a principle axis factor analysis and these were termed as urgency of restoring the level of nicotine after abstinence during sleeping and maintaining the level of nicotine during waking. The internal reliability (Cronbach's alpha coefficient) of the whole FTCD is low (.68) as well as for the two subscales (.60) and (.62) respectively. Conclusion: The psychometric properties of the Arabic version of the FTCD scale in this sample of Yemeni khat chewers who smoked regularly confirmed what has been established in other cultural settings. The findings of this study have yet to be cross validated amongst other appropriately representative sample of Arabic speakers.

• International Commerce and Information Review
• /
• v.16 no.2
• /
• pp.199-218
• /
• 2014

The study was focused on the Energy Efficiency of Korea. Thus this study calculates Energy Intensity and Energy Dependence, Energy Independence rate using data Korea Energy Statistics. This study calculate 'Energy Intensity', 'Energy Dependence', 'Energy Independence rate', which is making use of the Gross Domestic product, Energy Consumption, Energy Export and Import, Energy Gross Input. The analysis of the pater suggests that this goal can be achieved throuth improved efficiency of Energy consumption and Energy Independence.