http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
뇌졸중 발생 예측모형을 위한 Cox와 Weibull 모형의 비교 평가
김윤남(Youn Nam Kim),조어린(Ur Rin Cho),남병호(Byung-Ho Nam),박일수(Il Soo Park),지선하(Sun Ha Jee) 한국역학회 2008 Epidemiology and Health Vol.30 No.1
Objective: The objective was to compare Cox proportional hazards model and Weibull model for predicting long-term probabilities for stroke risk in the Korean Cancer Prevention Study(KCPS).<BR> Methods: The subjects comprised of 385,279 Korean aged 55 to 64 years who received health insurance from the National Health Insurance Corporation and who had medical examinations in 1992 and 1995. 70% of the subjects were used for model building and the rest for model evaluation. The final prediction model for stroke includes age, systolic blood pressure, diabetes, total cholesterol and smoking. Subjects were follow-up for identification of incident stroke cases between 1993 and 2005. Comparisons included predicted coefficients of stroke risk factors, incidence probabilities over 10 years, and the area under a receiver operating characteristics (ROC) curve for both Cox"s proportional hazard model and Weibull model.<BR> Results: The average age of study population was 55.5 years in men and 56.3 years in women, respectively. Percentage of men and women in study population were 58.0% and 42.0%, respectively. The study findings satisfied proportionality according to the two models. There was no significant difference in coefficients between the two models of prediction models in men and in women. Moreover, there was no difference in incidence probabilities of stroke and c-statistics. C-statistics were 0.68 for men as same as for women.<BR> Conclusion: There was no difference for the prediction of the stroke risk in the Korean population using Cox"s proportional hazard model and Weibull model, thus the two models were found to be efficient for this purpose.
종속적인 중도절단을 가진 동물종양 자료의 분석을 위한 모형
김진흠,김윤남,Kim, Jin-Heum,Kim, Youn-Nam 한국통계학회 2010 응용통계연구 Vol.23 No.5
동물종양 실험에서는 종양발생 시간이 직접 관찰되지 않고 단지 자연사로 인한 관찰 시점이나 강제적으로 희생시킨 시점 이전에 종양이 발생했는지 유무만을 알 수 있다. 이와 같은 형태의 결측을 가진 자료를 분석하기 위해 3단계(건강$\rightarrow$종양발생$\rightarrow$사망) 모형이 널리 사용되고 있다. 본 논문에서는 자연사로 인한 사망 시간이 종속적인 중도절단으로 작용하여 사망 시간과 종양발생 시간이 종속될 때, 이를 모형에 반영하기 위해 감마 프레일티 효과를 도입하였다. 모수 추정은 종양발생 시간과 프레일티 효과의 결측을 다루기 위해 EM 알고리즘 방법을 사용하였다. 제안한 추정량의 소표본 성질을 살펴보기 위해 제안한 방법을 Lindsey와 Ryan (1993, 1994)의 방광암 자료에 적용하여 모수를 추정하였으며, 그 추정값을 바탕으로 모의실험을 수행하였다. In animal tumorigenicity data, the occurrence time of tumor is not observed because the existence of a tumor is examined only at either time of natural death or time of sacrifice for the animal. A three-state model (Health-Tumor onset-Death) is widely used to model the incomplete data. In this paper, we employed a frailty effect into the three-state model to incorporate the dependency of death on tumor occurrence when the time of natural death works as an informative censoring against the tumor onset time. For the inference of parameters, then the EM algorithm is considered in order to deal with missing quantities of tumor onset time and random frailty. The proposed method is applied to the bladder tumor data taken from Lindsey and Ryan (1993, 1994) and a simulation study is performed to show the behavior of the proposed estimators.
군집의 크기가 생존시간에 영향을 미치는 군집 구간중도절단된 자료에 대한 준모수적 모형
김진흠,김윤남,Kim, Jinheum,Kim, Youn Nam 한국통계학회 2014 응용통계연구 Vol.27 No.2
본 논문에서는 군집 구간중도절단된 자료에서 생존시간이 군집의 크기에 의존할 때 주변모형으로부터 가중 추정 방법과 군집 내 재추출 방법을 써서 모수를 추정하고 그 추정량의 점근적 성질을 살펴보았다. 모의실험을 통해 추정량의 편향의 크기와 신뢰구간의 포함율 측면에서 볼 때 제안한 두 추정 방법이 생존시간과 군집의 크기 간의 종속 관계를 무시한 방법보다 우수한 것으로 나타났다. 제안한 추정 방법을 림프성 사상충 자료에 적용한 결과에 따르면 서로 다른 두 치료방법이 유의하게 다르지 않았으며 나이 효과도 매우 유의하지 않은 것으로 나타났다. We propose two estimating procedures to analyze clustered interval-censored data with an informative cluster size based on a marginal model and investigate their asymptotic properties. One is an extension of Cong et al. (2007) to interval-censored data and the other uses the within-cluster resampling method proposed by Hoffman et al. (2001). Simulation results imply that the proposed estimators have a better performance in terms of bias and coverage rate of true value than an estimator with no adjustment of informative cluster size when the cluster size is related with survival time. Finally, they are applied to lymphatic filariasis data adopted from Williamson et al. (2008).
반복 하중 하에서 에너지 기반 손상해석 기법을 적용한 재질열화된 CF8A 재료의 파괴인성 예측 기법
윤교근(Gyo-Geun Youn),남현석(Hyun-Suk Nam),전준영(Jun-Young Jeon),김윤재(Yun-Jae Kim),김진원(Jin-Won Kim) 대한기계학회 2017 대한기계학회 춘추학술대회 Vol.2017 No.11
This paper suggests a method to predict thermal ageing effect of CF8A material under cyclic loading effect. To achieve the goal, FE damage model based on multi-axial ductility theory is used. The model describes the local failure by fracture strain energy and stress-triaxiality. The fracture strain energy of aged CF8A can be determined from tensile test and that of aged CF8A can be assessed adopting thermal ageing constant “C”. To simulate the material behaviour under cyclic loading, combined hardening model is adopted and to predict fracture toughness under cyclic loading condition, two assumptions are made. One is about energy conservation and the other is about crack opening. From the assumptions, same fracture criteria are used to predict fracture toughness under cyclic loading condition.
조호진 ( Ho Jin Cho ),주우현 ( Woo Hyun Joo ),김윤남 ( Youn Nam Kim ),배종면 ( Jong Myon Bae ),남정모 ( Chung Mo Nam ) 한국보건정보통계학회(구 한국보건통계학회) 2014 보건정보통계학회지 Vol.39 No.2
Objectives: The purpose of the study is to review various methods in age-period-cohort (APC) analysis and to provide a guideline to choose adequate method evaluating age, period, and cohort effects. We investigated age, period, and cohort effects of breast cancer incidence between 1999 and 2011 in Korea. Methods: Data on female breast cancer incidence from 1999 to 2011 were drawn from the Korean national statistical office. The 5-year period of data units (1999-2003, 2004-2008, and 2009-2011) and 5-year age interval (30-34-80-84) were used to calculate 13 birth cohorts. The graphical approach, constrained generalized linear model (CGLM) approach, median polish approach and intrinsic estimator (IE) approach were used to estimate age, period, and cohort effects. Results: The age and period effects existed significantly in CGLM, median polish, IE approaches. The breast cancer incidence increased along with age and period. However, there was a difference in cohort effect. For CGLM, positive cohort effects for recent cohort emerged significantly, but for the other methods, no significant effects shown. Conclusions: While previous studies have used the CGLM method, CGLM depends on arbitrary parameter constraints. Therefore, we suggest median polish approach or IE approach for analyzing APC models to obtain more accurate results.