• The Korean Journal of Applied Statistics
• /
• v.27 no.1
• /
• pp.89-102
• /
• 2014

In this study, we analyze the effect on the smoking status in the Seoul Metropolitan area using a multilevel logistic model with Community Health Survey data from the Korea Centers for Disease Control and Prevention. Intraclass correlation coefficient (ICC), profiling analysis and two types of predicted value were used to determine the appropriate multilevel analysis level. Sensitivity, specificity, percentage of correctly classified observations (PCC) and ROC curve evaluated model performance. We showed the applicability for multilevel analysis allowed for the possibility that different factors contribute to within group and between group variability using survey data.

• Journal of Korean Society of Transportation
• /
• v.20 no.7
• /
• pp.79-85
• /
• 2002

In this paper, multilevel models are adopted to identify interactions among household members in trip making behaviors. The multilevel approach is a proper methodology to handle samples, which are extracted from a hierarchical structure universe. PSTP dataset is used in developing models and understand proportion of variations among individuals and household. The results of this study show that for activity participation and travel behavior household level variance is more than 1/4 of person level variance and therefore not negligible. The results confirm the importance of multilevel model in travel behavior analysis.

• Journal of Korean Society of Transportation
• /
• v.21 no.4
• /
• pp.145-154
• /
• 2003

Multilevel and Multivariate Structural Equation Mpdel is applied to handle the hierarchical nature of the data and explain complex relationship among socioeconomic factors of individuals and household, activity participation, and travel behavior using Puset Sound Transportation Panel data. From analysis, variations of individual activity participation and travel behavior can be divided into two categories : Within-household variation and Between-households variation. Empirical results show that the interdependency index(p) of variables for household members within a household is between 0.13 and 0.33 indicating high interdependency. These results suggest that Multilevel and Multivariate SEM approach is an appropriate modeling methodology and gives additional information for activity participation and travel behavior. Also most of personal and household characteristics influence on activity participation and travel behavior within a household as well as between households.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2006.05a
• /
• pp.110-116
• /
• 2006

지금까지 대부분의 리던던시 할당문제(RAP: redundancy allocation problems) 관련 연구들에서는 최상위 수준에서의 시스템 리던던시보다는 최하위 수준인 부품의 리던던시를 고려하였다. 이는 최하위 수준에서의 리던던시가 최상위 수준의 리던던시보다 효과적이라고 알려진 일반적 원리 때문이었다. 최근 한 연구에서는 동일하지 않은 예비부품을 사용하여 리던던시를 실시하는 경우 직렬구조의 시스템에서도 일반적 원리와 다른 결과가 나타날 수 있음을 보이고, 시스템을 구성하는 모든 수준에서 리던던시가 가능한 다수준 리던던시 할당문제(MRAP: multi-level RAP)를 제시하였다. 그러나 MRAP는 모든 수준에서의 리던던시를 고려하지만 단지 한 수준을 선택하여 리던던시를 할 수 있다는 가정사항을 포함하고 있다. 본 연구에서는 MRAP의 이러한 가정사항을 완화하여 시스템을 구성하는 모든 수준에서 리던던시를 위한 수준을 복수로 선택 가능한 혼합 다수준 리던던시 할당문제(MMRAP: multiple MRAP)를 제시하고 모형화하며, 문제의 해법을 위한 유전자 알고리듬(GA: genetic algorithm)을 제시한다. 제시한 GA를 활용한 몇 가지 수치실험을 통해 모형이 기존의 RAP 경우보다 효과적임을 입증한다.

• The Korean Journal of Applied Statistics
• /
• v.21 no.6
• /
• pp.923-932
• /
• 2008

We are likely to face complex multivariate data which can be characterized by having a non-trivial correlation structure. For instance, omitted covariates may simultaneously affect more than one count in clustered data; hence, the modeling of the correlation structure is important for the efficiency of the estimator and the computation of correct standard errors, i.e., valid inference. A standard way to insert dependence among counts is to assume that they share some common unobservable variables. For this assumption, we fitted correlated random effect models considering multilevel model. Estimation was carried out by adopting the semiparametric approach through a finite mixture EM algorithm without parametric assumptions upon the random coefficients distribution.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.2
• /
• pp.499-510
• /
• 2016

It is very important to select relevant variables in regression models for survival analysis. In this paper, we introduce a penalized variable-selection procedure in multi-level frailty models based on the "frailtyHL" R package (Ha et al., 2012). Here, the estimation procedure of models is based on the penalized hierarchical likelihood, and three penalty functions (LASSO, SCAD and HL) are considered. The proposed methods are illustrated with multi-country/multi-center bladder cancer survival data from the EORTC in Belgium. We compare the results of three variable-selection methods and discuss their advantages and disadvantages. In particular, the results of data analysis showed that the SCAD and HL methods select well important variables than in the LASSO method.

• Korean Journal of Social Welfare Studies
• /
• v.43 no.1
• /
• pp.389-411
• /
• 2012

A new era of research has focused on examining the growth of change in drinking problems among the elderly. Thus, the purpose of the present study was two fold: (1) to investigate trajectories of drinking problems(CAGE) among the Korean elderly(age$${\geq_-}65$$); and (2) to identify the predicting factors for the intercept and the slope of alcohol problems using multi-level growth curve model. Data come from three waves(1st wave(2006)~3rd wave(2008) of the Korea Welfare Panel(KWP) study. The results indicated that the levels of drinking problems decreased over time and that age, gender, marital status, religion, poverty, self-rated health, and social relationship satisfaction were associated with the baseline CAGE. Further analysis showed that social relationship satisfaction affected the declining slope of drinking problems over time. Specifically, among those who satisfied social relationship, there was a sharp decline of CAGE over time. Overall findings highlight the importance of developing and implementing effective alcohol prevention programs for the elderly in the community settings to mitigate the harmful effects of various psycho-social stressors. Especially, programs to maintain and form healthy social support network are suggested as critical interventions for prevention as well as recovery of alcohol problems in late life.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.5
• /
• pp.1169-1182
• /
• 2016

Cancer is a typical cause of death in Korea that becomes a major issue in health care. According to Cause of Death Statistics (2014) by National Statistical Office, SMRs (standardized mortality rates) in Busan were counted as the highest among all cities. In this paper, we used data of Busan Regional Cancer Center to estimate the extent of the cancer incidence rate and cancer mortality rate. The data are considered in small areas of administrative units such as Gu/Dong from years 2003 to 2009. All cancer including four major cancers (stomach cancer, colorectal cancer, lung cancer, liver cancer) have been analyzed. We carried out model selection and parameter estimation using spatial multi-level model incorporating a spatial correlation. For the spatial effects, CAR (conditional autoregressive model) has been assumed.

• Korean Journal of Labor Studies
• /
• v.19 no.1
• /
• pp.35-64
• /
• 2013

This paper explores whether or not a minimum wage increase can do much to alleviate working poor. For this purpose, I analyze transitions from working poor to working non-poor and to unemployment or non-economically active states, using KLIPS (Korea Labor and Income Panel Study). This study uses the multilevel multinomial logit model to control unobserved individual heterogenous characteristics. It finds that a minimum wage increase tends to cause a higher probability of transitions from working poor to working non-poor. It is also discovered that a minimum wage increase is not negatively related with the persistence of the working state. It is concluded that minimum wage increases are likely to be effective in improving the living standards of the 'working poor'.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.34 no.2
• /
• pp.561-571
• /
• 2014

This study analyzes traffic accident severity of old age drivers in fourteen cities and counties of Jeonbuk Province. It is assumed that traffic accident effecting factors have two staged structure by personal and driving environment and urban characteristics. Multilevel Analysis Model is used under the assumption of hierarchical characteristics to analyze factors effecting severity. As the driver's age increases after sixty-five years old, accident damages become severe. The drunk driving is likely to make traffic accident damage more severer. The number of fatal accident by old age drivers is about three time more than by no old age drivers. Old age drivers have higher number of night traffic accidents but severer ones in daytime. Old age drivers show the higher number of traffic accidents but severer ones in fine weather. Wet road surface also influences damage severity and especially old age drivers show higher serious damage and fatal than no old drivers.