• Communications for Statistical Applications and Methods
• /
• 제29권1호
• /
• pp.1-25
• /
• 2022

A modified Bagdonavičius-Nikulin chi-square goodness-of-fit is defined and studied. The lymphoma data is analyzed using the modified goodness-of-fit test statistic. Different non-Bayesian estimation methods under complete samples schemes are considered, discussed and compared such as the maximum likelihood least square estimation method, the Cramer-von Mises estimation method, the weighted least square estimation method, the left tail-Anderson Darling estimation method and the right tail Anderson Darling estimation method. Numerical simulation studies are performed for comparing these estimation methods. The potentiality of the new model is illustrated using three real data sets and compared with many other well-known generalizations.

• 대한교통학회지
• /
• 제25권5호
• /
• pp.135-148
• /
• 2007

교통사고의 특성과 사고에 대한 지속시간 사이의 관계에 대한 이해는 사고의 효과적인 대응과 사고에 의한 혼잡을 감소시키는데 핵심 요소가 된다. 때문에 본 연구의 목적은 AFT metric 모형을 적용한 사고 지속시간을 분석하는 것이다. 비록 로그 로지스틱 및 로그 정규 AFT 모형이 통계적 이론과 기존 연구 사례를 기반으로 선정되었으나, 로그 로지스틱 모형이 보다 우수하게 추정되었다. AFT 모형은 예측 목적으로도 널리 사용되기 때문에, 추정된 모형은 사고 발생시 사고 관련 기본 정보 접수 즉시 고속도에서의 사고 지속시간 예측에 사용될 수 있다. 결과적으로, 예측된 사고 지속시간 정보는 사고를 처리하기 위한 제반 의사 결정에 도움을 줄 뿐 아니라 교통 혼잡의 감소 및 추가 사상자의 감소로 그 효과가 이어질 것으로 판단된다.

• 응용통계연구
• /
• 제28권6호
• /
• pp.1191-1208
• /
• 2015

대부분의 통계분석방법은 요약통계량에 의존하지만 그래픽적 방법을 이용하면 자료의 특성을 파악하기 쉽고 통계량만으로는 알아낼 수 없는 부분까지도 접근이 가능하다. 그래프를 통한 로지스틱회귀모형의 평가 방법으로 로그-밀도비를 통한 검토, 차원 검토, 주변모형산점도, 카이잔차산점도, CERES 그림을 알아보고 모의자료들을 통해 다양한 상황에서 그래픽적 방법들 어떠한 결과를 나타내지를 비교 검토한다.

• 한국축산식품학회지
• /
• 제32권1호
• /
• pp.62-67
• /
• 2012

This study calculated kinetic parameters of Escherichia coli O157:H7 and developed a probabilistic model to estimate growth probabilities of E. coli O157:H7 on polyethylene cutting boards as a function of temperature and time. The surfaces of polyethylene coupons ($3{\times}5$ cm) were inoculated with E. coli O157:H7 NCCP11142 at 4 Log $CFU/cm^2$. The coupons were stored at 13 to $35^{\circ}C$ for 12 h, and cell counts of E. coli O157:H7 were enumerated on McConkey II with sorbitol agar every 2 h. Kinetic parameters (maximum specific growth rate, Log $CFU/cm^2/h$; lag phase duration, h; lower asymptote, Log $CFU/cm^2$; upper asymptote, Log $CFU/cm^2$) were calculated with the modified Gompertz model. Of 56 combinations (temperature${\times}$time), the combinations that had ${\geq}$0.5 Log $CFU/cm^2$ of bacterial growth were designated with the value of 1, and the combinations that had increases of <0.5 Log $CFU/cm^2$ were given the value 0. These growth response data were fitted to the logistic regression to develop the model predicting probabilities of E. coli O157:H7 growth. Specific growth rate and growth data showed that E. coli O157:H7 cells were grown at $28-35^{\circ}C$, but there were no obvious growth of the pathogen below $25^{\circ}C$. Moreover, the developed probabilistic model showed acceptable performance to calculate growth probability of E. coli O157:H7. Therefore, the results should be useful in determining upper limits of working temperature and time, inhibiting E. coli O157:H7 growth on polyethylene cutting board.

• Journal of the Korean Data and Information Science Society
• /
• 제25권1호
• /
• pp.177-185
• /
• 2014

로지스틱회귀모형에서 설명변수만으로는 충분히 설명이 되지 못하고 설명변수의 변환된 형태인 이차항 또는 교호작용항이 필요한 경우가 있다. 설명변수가 두 개이고 조건부 분포가 이변량 정규분포를 따르는 경우 로지스틱회귀모형에서는 기본적으로 이차항과 교호작용항이 모형에 포함되어야 한다. 하지만 조건부 분포의 분산과 상관계수에 따라 이차항과 교호작용항이 필요하지 않게 되는 경우도 있다. 분산이나 상관계수에 대한 정보는 산점도를 보고 대체적인 판단이 가능하지만 교호작용항의 필요성을 판단하기가 쉽지 않다. 본 논문에서는 3차원 잔차산점도를 이용한 교호작용의 탐색방법을 제시하고 이 방법을 실제 자료에 적용시켜본다.

• 응용통계연구
• /
• 제18권3호
• /
• pp.701-712
• /
• 2005

로지스틱 회귀모형에서 suppression의 논의는 선형회귀의 논의보다 많지 않은데 그 이유 중의 하나는 회귀제곱합 또는 결정계수의 정의가 유일하지 않고 다양하기 때문이다. 여러 종류의 결정계수들 중에서 선호되는 두 종류의 결정계수와 Liao와 McGee(2003)가 제안한 두 종류의 수정 결정계수의 정의로부터 회귀제곱합을 유도하여 로지스틱 회귀모형에서의 suppression을 설명하고자 한다. 모의실험을 통하여 자료를 생성하여 어떤 경우에 suppression이 발생하는지를 살펴보고 그 결과를 선형회귀모형에서의 suppression 결과와 비교한다.

• Journal of Advanced Research in Ocean Engineering
• /
• 제4권4호
• /
• pp.185-194
• /
• 2018

In this study, the extreme sloshing pressure was predicted using various statistical models: three-parameter Weibull distribution, generalized Pareto distribution, generalized extreme value distribution, and three-parameter log-logistic distribution. The estimation of sloshing impact pressure is important in design of liquid cargo tank in severe sea state. In order to get the extreme values of local impact pressures, a lot of model tests have been carried out and statistical analysis has been performed. Three-parameter Weibull distribution and generalized Pareto distribution are widely used as the statistical analysis method in sloshing phenomenon, but generalized extreme value distribution and three-parameter log-logistic distribution are added in this study. Additionally, statistical distributions are fitted to peak pressure data using three different parameter estimation methods. The data were obtained from a three-dimensional sloshing model text conducted at Seoul National University. The loading conditions were 20%, 50%, and 95% of tank height, and the analysis was performed based on the measured impact pressure on four significant panels with large sloshing impacts. These fittings were compared by observing probability of exceedance diagrams and probability plot correlation coefficient test for goodness-of-fit.

• 한국환경보건학회지
• /
• 제44권4호
• /
• pp.380-390
• /
• 2018

Objective: A significant association between blood lead levels and hypertension has been reported in many studies. The relationship between cadmium and hypertension has been debated as well. We aimed to study the association of lead, cadmium, and both with hypertension in the Korean general population. Methods: We examined 5,967 adult men and 6,074 women who participated in the Korea National Health and Nutrition Examination Survey III-VI (2005, 2008-2013 years). Logistic regression models were used to examine the relationship between blood lead concentration and blood cadmium concentration and hypertension using logtransformed blood lead and cadmium concentrations as independent variables after covariate adjustment. Results: Adjusted for general characteristics, the odds ratio of log-lead to hypertension was 2.71 (1.82-4.03), and log-cadmium to hypertension was 2.52 (1.83-3.47). Estimates were found to be statistically significant (p<0.001). When a multiple logistic model was applied, the odds ratio of log-lead and log-cadmium for hypertension were 2.24 (1.50-3.36) and 2.24 (1.62-3.10), respectively. The standardized estimate coefficients of log-lead and logcadmium for hypertension were 4.77 and 6.65, respectively. Conclusion: We observed the association of blood lead concentration, blood cadmium concentration, and both with hypertension. This study suggests that exposure to lead and exposure to cadmium are both risk factors for hypertension.

• 한국전자통신학회논문지
• /
• 제18권5호
• /
• pp.867-876
• /
• 2023

본 연구에서는 감마족 분포(Erlang, Log-Logistic, Rayleigh)을 적용한 NHPP 소프트웨어 개발 비용 모형의 속성을 새롭게 분석하였고, 모형의 속성을 검증하기 위해 Goel-Okumoto 기본 모형과 비교한 후, 이를 근거로 최적의 모형도 제시하였다. 소프트웨어 신뢰도를 분석하기 위하여 시스템 운영 중 랜덤하게 발생한 고장 시간 데이터를 활용하였고, 모수의 계산은 최우추정법을 사용하여 해결하였다. 다양한 속성 분석(평균값 함수, 개발 비용, 최적의 방출시간)을 통하여 종합적으로 평가한 결과, Rayleigh 모형이 가장 우수한 성능을 가진 모형임을 확인하였다. 본 연구를 통하여, 기존 연구 사례가 없는 감마족 분포를 적용한 소프트웨어 개발비용 모형의 속성을 새롭게 규명하였다. 또한, 개발자들이 초기 단계에서 본 연구 데이터를 효율적으로 활용할 수 있도록 기초적인 설계 데이터도 제시할 수 있었다.

• 수산경영론집
• /
• 제46권3호
• /
• pp.103-118
• /
• 2015

The purpose of this paper is to discuss possibilities of applying growth curve models, such as Logistic, Log-Logistic, Log-Normal, Gompertz and Weibull, to three specific technology areas of Fishery Science and Technology in the process of measuring their technology level between Korea and countries with the state-of-the art level. Technology areas of hazard control of organism, environment restoration, and fish cluster detect were selected for this study. Expert panel survey was conducted to construct relevant panel data for years of 2013, 2016, and a future time of approaching the theoretical maximum technology level. The size of data was 70, 70 and 40 respectively. First finding is that estimation of shape and location parameters of each model was statistically significant, and lack-of-fit test using estimated parameters was statistically rejected for each model, meaning all models were good enough to apply for measuring technology levels. Second, three models other than Pearl and Gompertz seemed very appropriate to apply despite the fact that previous case studies have used only Gompertz and Pearl. This study suggests that Weibull model would be a very valid candidate for the purpose. Third, fish cluster detect technology level is relatively higher for both Korea and a country with the state-of-the-art among three areas as of 2013. However, all three areas seem to be approaching their limits(highest technology level point) until 2020 for countries with the state-of-the-art. This implies that Korea might have to speed up her research activities in order to catch up them prior to 2020. Final suggestion is that future study may better apply various and more appropriate models respectively considering each technology characteristics and other factors.