Measuring the performances of covariates using exponential survival analysis with partly-interval censored simulation data

In many fields of science, modelling and analyzing survival rates has shown to be a valuable element of statistical study. This paper aims at proposing the partly-interval censored data into the fixed and time-varying covariates and measure the performances of Exponential survival distribution using...

Full description

Bibliographic Details
Published in:AIP Conference Proceedings
Main Author: Jamil S.A.M.; Lai J.; Abdullah M.A.A.
Format: Conference paper
Language:English
Published: American Institute of Physics 2024
Online Access:https://www.scopus.com/inward/record.uri?eid=2-s2.0-85188429036&doi=10.1063%2f5.0194223&partnerID=40&md5=618c17d4698c33d26ebfc696d7c1b91b
Description
Summary:In many fields of science, modelling and analyzing survival rates has shown to be a valuable element of statistical study. This paper aims at proposing the partly-interval censored data into the fixed and time-varying covariates and measure the performances of Exponential survival distribution using mean square error (MSE), mean bias error (MBE), mean absolute error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE) and standard error. As a result, when dealing the data without censored observations, the exponential distribution significantly fit the simulation data since low values of error measurements appeared when the data included the exact and complete types of simulation. Thus, this study proposed that the uncensored data could be applicable towards the Exponential survival distribution compared to other distributions of survival analysis. © 2024 Author(s).
ISSN:0094243X
DOI:10.1063/5.0194223