PERBANDINGAN ANTARA METODE BAYESIAN SELF PRIOR GAMMA DAN PRIOR JEFFREY DALAM MENGESTIMASI PARAMETER MODEL SURVIVAL BERDISTRIBUSI WEIBULL DATA TERSENSOR

Nurwakia, Nurwakia (2020) PERBANDINGAN ANTARA METODE BAYESIAN SELF PRIOR GAMMA DAN PRIOR JEFFREY DALAM MENGESTIMASI PARAMETER MODEL SURVIVAL BERDISTRIBUSI WEIBULL DATA TERSENSOR. S2 thesis, UNIVERSITAS NEGERI MAKASSAR.

[img] Text
Artikel Prior Gamma dan Prior Jeffrey.pdf
Download (753kB)

Abstract

This study is an applied research that aims to compare Bayesian SELF Prior Gamma and Prior Jeffrey from the estimated survival model with censored Weibull data. The data used are secondary data from the medical recap of length of stay of patients with DHF at the Hajj Hospital (RS Haji) in Makassar City from 2015 to 2019 with 275 data, consisted of 34 censored data and 241 uncensored data. The survival data is the data that shows the time an individual can survive until a certain event occurs. Survival data is said to be censored if the object in the study is lost or until the end of the study the object has not experienced a certain event. The test indicators of this study are the minimum value of Mean Square Error (MSE). The MSE values obtained for the Bayesian SELF Prior Gamma estimation results are 0.0225 and 1.2291, the Bayesian SELF Prior Jeffrey are 0.0107 and 1.0810. Based on the MSE value obtained, the Bayesian SELF Prior Jeffrey method is better than the Bayesian SELF prior Gamma method in estimating the Weibull distribution survival model parameters with censored data. The estimation results obtained the longer the patient's stay, the better the recovery rate. The results of this study can provide information for the hospital in handling and providing treatment for patients with DBD. Keywords: Weibull Distribution, Bayesian SELF Method, Prior Gamma, Prior Jeffrey

Item Type: Thesis (S2)
Subjects: PASCASARJANA
PASCASARJANA > MATEMATIKA
Divisions: PROGRAM PASCASARJANA
Depositing User: Dede Yulistian A.Md
Date Deposited: 26 Mar 2021 05:59
Last Modified: 26 Mar 2021 05:59
URI: http://eprints.unm.ac.id/id/eprint/18648

Actions (login required)

View Item View Item
Test Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.