The logistic regression analysis with nonparametric approach based on Local scoring algorithm (case study: Diabetes Mellitus Type II cases in Surabaya of Indonesia)

One of the statistical methods used to analyze the relationship between categorical scale response variable and categorical or continuous variable of predictors is logistic regression analysis. There are two ways of approaching regression model. The global approach assumes that the regression model for each individual observation has the same parameters, whereas the local approach assumes not all individuals have the same parameters. Regression model with local approach is often called nonparametric regression. The purpose of this research will be to develop the risk model of Diabetes Mellitus Type II incidence in patients of The Hajj General Hospital of Surabaya with nonparametric logistic regression approach by the local scoring algorithm. Based on the result of this research concluded that the probability of a person's risk of Diabetes Mellitus Type II increases with age up to about 60 years and after that tends to decrease, the more the increase in one's BMI there is a significant increase in the prevalence of Diabetes Mellitus Type II. The validity of the estimation of this obtained model is 88.89%.