MODEL REGRESI DATA TAHAN HIDUP TERSENSOR TIPE III BERDISTRIBUSI LOG-LOGISTIK

Abstract

Lifetime T is the time from initial treatment until the first response is to be observed which can be death due to a particular disease, illness that recur after treatment or the emergence of new diseases. In research of survival testing the data term are censored and not censored. Censored observation occur if the survival time of the observed individual is not known with certainty while the observation not censored if the survival time of observation is known with certainty. There are three different types of censoring observation that are type I,type II, and type III. Censoring type III is an observation made to several individuals at different time within a certain period, this is because an individual entry into the observations at different times. Influence of other factros on the response variable that is survival time relation should be considered. One way to know relationship is through a regression model. Regression model of survival data with censoring type III of log-logistic distribution is made following the curve of the response variable. Estimation of parameters using maximum likelihood methods. Regression model was apllied to estimate the survival time of patients with lung cancer for factors of the infected cell and type of treatment.