• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.431-434
• /
• 2003

In this paper we introduce a modified objective function for fuzzy c-means clustering with logistic regression model in the presence of noise cluster. The logistic regression model is commonly used to describe the effect of one or several explanatory variables on a binary response variable. In real application there is very often no sharp boundary between clusters so that fuzzy clustering is often better suited for the data.

• Communications for Statistical Applications and Methods
• /
• v.13 no.2
• /
• pp.369-378
• /
• 2006

This Study focuses on the binary forecast of occurrence of heavy snow in Honam area based on the MOS(model output statistic) method. For our study daily amount of snow cover at 17 stations during the cold season (November to March) in 2001 to 2005 and Corresponding 45 RDAPS outputs are used. Logistic regression model and neural networks are applied to predict the probability of occurrence of Heavy snow. Based on the distribution of estimated probabilities, optimal thresholds are determined via true shill score. According to the results of comparison the logistic regression model is recommended.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.6
• /
• pp.157-165
• /
• 2016

Among high-tech products, the mobile phone has experienced a rapid rate of innovation and a shortening of its product life cycle. The shortened product life cycle poses major challenges to those involved in the creation of forecasting methods fundamental to strategic management and planning systems. This study examined whether the best model applies to the entire diffusion life span of a mobile phone. Mobile phone sales data from a specific mobile service provider in Korea from March of 2013 to August of 2014 were analyzed to compare the performance of two S-shaped diffusion models and two non-linear regression models, the Gompertz, logistic, Michaelis-Menten, and logarithmic models. The experimental results indicated that the logistic model outperforms the other three models over the fitted region of the diffusion. For forecasting, the logistic model outperformed the Gompertz model for the period prior to diffusion saturation, whereas the Gompertz model was superior after saturation approaches. This analysis may help those estimate the potential mobile phone market size and perform inventory and order management of mobile phones.

• Journal of the Korean Data and Information Science Society
• /
• v.15 no.2
• /
• pp.367-378
• /
• 2004

Recently logistic regression is popular in a variety of fields so that a number of statistical packages are developed for analyzing the logistic regression. This paper briefly considers the several types of logistic regression models used depending on different types of data. In addition, when four statistical packages (MINTAB, SAS, SPSS and STATA) are used to apply logistic regression models to the real fields respectively, their scope and characteristics are investigated.

• The Korean Journal of Applied Statistics
• /
• v.22 no.1
• /
• pp.153-161
• /
• 2009

We consider a logistic model with random effects as the superpopulation for estimating the small area pro-portions. The best linear unbiased predictor under linear mired model is popular in small area estimation. We use this type of estimator under logistic mixed motel for the small area proportions, on which the estimation of mean squared error is also discussed. Two kinds of estimation methods, the parametric bootstrap and the linear approximation will be compared through a Monte Carlo study in the respects of the normality assumption on the random effects distribution and also the magnitude of sample sizes on the approximation.

• Journal of Korean Institute of Industrial Engineers
• /
• v.27 no.4
• /
• pp.366-373
• /
• 2001

Economic component screening procedures for multi-component assembly products supplied with warranty are proposed. It is assumed that the screening variable is continuous and the performance variable is dichotomous. Logistic and normal models are considered; the proportion of conforming items is a logistic function of the screening variable in the logistic model and the screening variable given the performance variable is normally distributed in the normal model. Cost models are constructed which involve three cost components; screening inspection cost for each component, cost due to disposing a rejected component, and warranty cost for an assembly product. Methods of finding the optimal screening procedures are presented and numerical examples are given.

• The Journal of the Korean dental association
• /
• v.44 no.10 s.449
• /
• pp.658-670
• /
• 2006

1980년대 후반기부터 치주질환과 뇌경색(ischemic stroke)자료의 연관성을 모색하는 시도가 있어왔다. 이번 연구의 목적은 치주질환과 뇌경색 유무와의 어떤 관계가 있는지를 60세 이상의 노인을 대상으로 조사, 통계 분석하였다. 자료는 미국의 총 국민조사 격인 The Third Nation Health and Nutrition Examination Survey (NHANES III)를 이용하였다. 이번 연구에서 unadjusted logistic model 통계법을 이용하여 치아 상실수와 뇌경색 경험이 통계학적으로 유의한 수치의 상관성이 있음을 알게 되었다. 또한 나이와 흡연유무를 고려, 조정한 후 multiple logistic model 통계법으로 잔존치아가 적을수록 더욱 뇌경색에 걸릴 확률이 높음을 보였다. 그러나 두 질병에 동시에 선택된 중요한 위험인자 (risk factor)를 모두 고려, 조정 한 후에는 통계학적인 유의성을 찾지 못했다. 치은퇴축, 치주낭 깊이, 치석, 탐침시 출혈과 뇌경색 경험은 각각의 비교법에서 약간의 상관성을 보이나, 모든 통계법을 통해 일괄된 결과를 얻을 수는 없었다.

• Journal of the Korean Statistical Society
• /
• v.36 no.1
• /
• pp.111-127
• /
• 2007

The logistic distribution is generalized using the Marshall-Olkin scheme and its generalization. Some properties are studied. First order autoregressive time series model with Marshall-Olkin semi-logistic distribution as marginal is developed and studied.

• Journal of the Korean Data Analysis Society
• /
• v.20 no.6
• /
• pp.2721-2731
• /
• 2018

This study purposes to derive the predictive model for the cold tolerance of Brassica napus, using the data collected in the Tree Breeding Lab of Gyeongsang National University during July and August of 2016. Three Brassica napus samples were treated at each of low temperatures from $4^{\circ}C$ to $-12^{\circ}C$ by decrement of $4^{\circ}C$, step by step, and electrolyte leakage levels were measured at each stage. Electrolyte leakages were observed tangibly from $-4^{\circ}C$. We tried to fit the six nonlinear regression models to the electrolyte leakage data of Brassica napus: 3-parameter logistic model, baseline logistic model, 4-parameter logistic model, (4-1)-parameter logistic model, 3-parameter Gompertz model, and (3-1)-parameter Gompertz model. The baseline levels of the electrolyte leakage estimated by these models were 4.81%, 4.07%, 4.19%, 4.07%, 4.55%, and 0%, respectively. The estimated median lethal temperature, LT50, were $-5.87^{\circ}C$, $-6.31^{\circ}C$, $-6.05^{\circ}C$, $-6.35^{\circ}C$, $-4.98^{\circ}C$, and $-5.15^{\circ}C$, respectively. We compared and discussed the measures of goodness of fit to select the appropriate nonlinear regression model.

• Clinical Nutrition Research
• /
• v.12 no.2
• /
• pp.138-153
• /
• 2023

The prevalence of metabolic syndrome (MetS) and its cost are increasing due to lifestyle changes and aging. This study aimed to develop a deep neural network model for prediction and classification of MetS according to nutrient intake and other MetS-related factors. This study included 17,848 individuals aged 40-69 years from the Korea National Health and Nutrition Examination Survey (2013-2018). We set MetS (3-5 risk factors present) as the dependent variable and 52 MetS-related factors and nutrient intake variables as independent variables in a regression analysis. The analysis compared and analyzed model accuracy, precision and recall by conventional logistic regression, machine learning-based logistic regression and deep learning. The accuracy of train data was 81.2089, and the accuracy of test data was 81.1485 in a MetS classification and prediction model developed in this study. These accuracies were higher than those obtained by conventional logistic regression or machine learning-based logistic regression. Precision, recall, and F1-score also showed the high accuracy in the deep learning model. Blood alanine aminotransferase (β = 12.2035) level showed the highest regression coefficient followed by blood aspartate aminotransferase (β = 11.771) level, waist circumference (β = 10.8555), body mass index (β = 10.3842), and blood glycated hemoglobin (β = 10.1802) level. Fats (cholesterol [β = -2.0545] and saturated fatty acid [β = -2.0483]) showed high regression coefficients among nutrient intakes. The deep learning model for classification and prediction on MetS showed a higher accuracy than conventional logistic regression or machine learning-based logistic regression.