• The Korean Journal of Applied Statistics
• /
• v.37 no.2
• /
• pp.225-237
• /
• 2024

In the analysis of competing risks data, some of covariates may not be fully observed for some subjects. In such cases, excluding subjects with missing covariate values from the analysis may result in biased estimates and loss of efficiency. In this paper, we studied multiple imputation and the augmented inverse probability weighting method for regression parameter estimation in the cause-specific proportional hazards model with missing covariates. The performance of estimators obtained from multiple imputation and the augmented inverse probability weighting method is evaluated by simulation studies, which show that those methods perform well. Multiple imputation and the augmented inverse probability weighting method were applied to investigate significant risk factors for the risk of death from breast cancer and from other causes for breast cancer data with missing values for tumor size obtained from the Prostate, Lung, Colorectal, and Ovarian Cancer Screen Trial Study. Under the cause-specific proportional hazards model, the methods show that race, marital status, stage, grade, and tumor size are significant risk factors for breast cancer mortality, and stage has the greatest effect on increasing the risk of breast cancer death. Age at diagnosis and tumor size have significant effects on increasing the risk of other-cause death.

• The Journal of Asian Finance, Economics and Business
• /
• v.5 no.3
• /
• pp.31-41
• /
• 2018

The study examines economic and environmental impacts of mass tourism on regional tourism destinations, particularly the establishment of "Ten New Bali", in Indonesia. The sample is restricted to the period of time in which annual data is available and comparable among variables from 1980 to 2015 (36 observations). All of the time series data was collected and retrieved from the World Development Indicator database published by the World Bank. This study applies cointegrating regression analysis using the fully modified OLS, canonical cointegrating regression, and dynamic OLS. The results of the study suggest that 1) there is a long-run equilibrium relationship between tourism receipts, environmental degradation and economic growth in Indonesia, 2) tourism growth and agriculture land growth are positively related to an increase of total output in the short-run in Indonesia, and 3) arable land is significant at the 0.01 level, but forest rents and CO2 from transport are not significant in the short-run in Indonesia. The results confirm that arable land is negatively related to an increase of total output in Indonesia. That is, when tourism growth in the economy is getting realized it shows that the environmental degradation increases greatly in inverse in the model, eventually negative impacts to the environment.

• Tribology and Lubricants
• /
• v.30 no.1
• /
• pp.15-20
• /
• 2014

A more accurate expression for effective viscosity is obtained using a linear regression of the data from Fukui-Kaneko's model, which are obtained through numerical calculations based on the linearized Boltzmann equation. Veijola and Turowski's expression is adopted as a base function for effective viscosity. The four coefficients in that equation are optimized, and sensitivity analysis is conducted for these coefficients. The results show that the coefficient for the first-order Knudsen number is the most accurate, whereas the coefficient in the exponential of the Knudsen number is the least accurate compared with Fukui-Kaneko's results. The expression for effective viscosity is accurate within 0.02% rms of Fukui-Kaneko's results for the inverse Knudsen numbers from 0.01 to 100 and surface accommodation coefficients ranging from 0.7 to 1.

• Journal of Korean Academy of Fundamentals of Nursing
• /
• v.6 no.3
• /
• pp.369-381
• /
• 1999

The purpose of this study was to determine the predictors of health promoting lifestyles(Health Promoting Lifestyle Profile HPLP) of Koreans in the Philippines. The sample consisted of 100 Koreans who have lived in Manila, Philippines. Data were collected for two months from Feb, 1 to March 30, 1998. Analysis of the data was done by use of percentage, t-test, ANOVA, Pearson's Correlation Coefficients, and Stepwise Multiple Regression. The results of this study were summarized as follows : 1. The range of total HPLP score was from 73 to 175 and the mean score of that was 117.23. In comparison of mean scores depending on each item of six dimensions, self-actualization dimension tended to show the hightest score(2.98) and responsibility for health dimension, the lowest score(1.83). 2. The HPLP score showed a positive correlation with age(r=.19), social support (r=.39), self-efficacy(r=.52) and perceived health status(r=.27), but a inverse correlation with perceived health care service utilization barriers(r=-.20). 3. By using stepwise multiple regression analysis it was determined that the main influencing factors on the HPLP score were self-efficacy(27.2%), social support(5.3%), age(5.3%) and perceived health status(2.9%). These variables made it possible to explain 40.7% of variance in HPLP score.

• Journal of Korean Academy of Fundamentals of Nursing
• /
• v.5 no.2
• /
• pp.353-361
• /
• 1998

The purpose of this study was to determine the predictors of health promoting behavior in postpartal $4{\sim}6$ week women. The sample consisted of 104 postpartal women who experienced a vaginal delivery at four obstetrical clinics located in Kwangju city. Data were collected for two months from June 1 to July 30, 1997. Analysis of the data was done by use of percentage, t-test, ANOVA, Pearson's Correlation Coeficients, and Stepwise Multiple Regression. The results of this study were summarized as follows : 1. The range of total HPLP score was from 93 to 182 and the mean score of that was 142.28. In comparison of mean scores defending on each item of six demensions, self-actualization demension tended to showed hightest score(3.33) and exercise & rest demension, the lowest score(2.47). 2. The HPLP score was not significantly different defending on the general characteristics. 3. The HPLP score showed a positive correlation with family-support(r=.51) and self-efficacy(r=.41), but a inverse correlation with perceived barriers(r=-.27). 4. By using stepwise multiple regression analysis it was determined that the main influencing factors on the HPLP score were family-support(26%) and self-efficacy (8%). These variables made it possible to explain 34% of variance in HPLP score.

• Korea journal of population studies
• /
• v.9 no.2
• /
• pp.42-52
• /
• 1986

An attempt had been made to obtain current information on induced abortion among currently married women aged 15 to 44. The source of data was 1985 National Fertility and Family Planning Survey conducted by Korea Institute for Population and Health in may 1985. 1. At the time of the survey, 53 percent of currently married women aged 15 to 44 had experienced induced abortion: 25.5 percent of the respondents who had experienced induced abortion only one time and another 27.6 percent more than two times. 2. The proportion of women who had experienced induced abortion seemed to increase according to the advance in age. It seemed that there was no significant difference in the experience rate of induced abortion by academic career, and there was inverse relationship between the experience rate of induced abortion and the age at first marriage. The experience rate if induced abortion by the number of living children was highest among those who had three children as 65.6 percent. 3. To analyze the effect of induced abortion on fertility, this study employed multiple regression analysis as a statistical technique. Instead of index representing fertility level the number of living children served as dependent variable. Independent variables used in analysis included age, age at first marriage, education level, ideal number of children, religion, frequency of induced abortion, total number of pregnancies and participation in labor force. Standardized partial regression coefficient of induced abortion was no less than -0.61. So, it can be concluded that induced abortion offered the great contribution on the birth control. A consistant health education and efficient management of family planning program would be essential for effectiveness of contraceptive practice.

• Journal of the Korean Society of Food Culture
• /
• v.34 no.2
• /
• pp.159-167
• /
• 2019

The purpose of this study is to provide basic information to improve understanding of contemporary kimchi culture in Korea. Many Koreans are now purchasing kimchi at markets, while the proportion of self-preparation is gradually decreasing. This commodification tendency of kimchi is considered to be associated with changes in consumer's behavior and attitude. In this study, a linear regression and a logistic regression model were used to identify relationships between kimchi consumption behavior and household characteristics. The results showed that the probability of kimjang activity was positively related with family size, possession of a kimchi refrigerator, self-preparation practice, and the intensity of sharing behavior. I also found that kimchi consumption volume per capita of 'purchasing' household was greater than that of 'self-preparing ' or 'sharing-dependent' households, and that the number of family members was inversely related with kimchi consumption volume per capita. The inverse relationship between family size and kimchi consumption volume per capita is considered to be contrary to the widespread thoughts in Korea, which have been developed while experiencing kimchi preparation and consumption in traditional extended families. I think that the relationship comes from differences in menu varieties, which appear to vary with family size. This issue will be investigated in subsequent studies.

• The Journal of Asian Finance, Economics and Business
• /
• v.8 no.2
• /
• pp.135-142
• /
• 2021

This study aims to examine the impact of audit committee characteristics on income management of companies listed on the Stock Exchange of Vietnam. Research data was collected from all 745 listed companies on Vietnam's stock market over four years, from 2015 to 2018. After excluding companies that did not qualify, there were 216 companies with 864 observations. With the help of dedicated software Stata 15, the impact of audit committee characteristics (through independent variables and control variables such as Audit Committee Independence, Auditing Committee size, Auditing Committee Expertise, Auditing Committee Meeting Frequency, Company Size, Financial Leverage, and Operating Cash Flow) to earning management through a multivariate regression model was determined. Research results from Vietnamese listed companies during this period show that the size and expertise of the audit committee are inversely related to the discretionary accruals representing earning management. At the same time, the research results also identify a positive relationship between firm size and earning management, and the inverse relationship between financial leverage, net cash flow from operating operations and earning management. However, the multivariate regression results do not find clear evidence of a relationship between audit committee independence and the audit committee meeting frequency to earning management.

• The Journal of Industrial Distribution & Business
• /
• v.12 no.6
• /
• pp.7-21
• /
• 2021

Purpose: Modern globalization and Western markets saturation has catalyzed the growth of culinary globalization into developing countries. The question was whether fast food consumers in frontier markets of Sub-Saharan Africa (Zimbabwe), either upholds national gastronomic tendencies in terms of consumer ethnocentrism and buy local or they adopt global fast food brands. Demographic consumer profiles were also analyzed as antecedents of consumer ethnocentrism. Research design, data and methodology: A sample size of 400 fast food-adult consumers was surveyed in the City of Harare. Data was captured on SPSS and Analysis of Moment Structure (AMOS). Hypothesis testing was done using sample t test (H₁), logistic regression (H₂) and multiple regression (H₃, ₄, ₅) analysis. Results: Consumer ethnocentrism in Zimbabwe was marginally above average and no statistically significant relationship between the levels of consumer ethnocentrism and adoption of foreign fast food brands was noted. Age had an inverse relationship; income had a positive association whilst gender had no statistical significance with consumer ethnocentrism. Conclusions: Despite the Zimbabwean consumers being marginally ethnocentric, international restaurateurs should invest in the Zimbabwean fast food market since their nature of being foreign has got an exotic appeal to the Zimbabwean consumers thereby enhancing their likelihood of success.

• The Korean Journal of Applied Statistics
• /
• v.33 no.1
• /
• pp.25-46
• /
• 2020

This paper presents ordinal probit semiparametric regression models using Bayesian Spectral Analysis Regression (BSAR) method. Ordinal probit regression is a way of modeling ordinal responses - usually more than two categories - by connecting the probability of falling into each category explained by a combination of available covariates using a probit (an inverse function of normal cumulative distribution function) link. The Bayesian probit model facilitates posterior sampling by bringing a latent variable following normal distribution, therefore, the responses are categorized by the cut-off points according to values of latent variables. In this paper, we extend the latent variable approach to a semiparametric model for the Bayesian ordinal probit regression with nonparametric functions using a spectral representation of Gaussian processes based BSAR method. The latent variable is decomposed into a parametric component and a nonparametric component with or without a shape constraint for modeling ordinal responses and predicting outcomes more flexibly. We illustrate the proposed methods with simulation studies in comparison with existing methods and real data analysis applied to a Korean National Health and Nutrition Examination Survey (KNHANES) 2016 for investigating nonparametric relationship between smoking behavior and coffee intake.