REGRESI LOGISTIK POLIKOTOMUS DENGAN DATA HILANG; POLYCHOTOMOUS LOGISTIC MODEL WITH MISSING VALUES

Suswanti, Sri Haryatmi

2012 | Disertasi | PROGRAM STUDI S2 MATEMATIKA

Logistic regression is an analytical tool generally applied in health studies and researchs whom the response variable have two values “success/yes” and “failure/no”. This paper intends to study polychotomous logistic regression models where the response variable has more than two categories and the covariates have missing values. In many medical data sets, we may face some missingness in some covariates such as denying to respond, lack of information in files, and incompleteness of study frame. In such case we deal with missing values. In this study, it is assumed that the missingness is at random and independent of deal with missing values. To obtain the estimator of parameters of the polychotomous logistic regression models it can use some method. In this thesis we use a Maximum Likelihood Estimation (MLE) of the parameter ? from an polychotomous logistic regression models. It has been seen that the estimators obtained are not available in nice closed form, so they can be easily evaluated by using Newton- Raphson solution method

Kata Kunci : Model Regresi Logistik Polikotomus; Estimasi Maksimum Likelihood; Metode Newton-Raphson; Data Hilang secara Acak