Waiting Time to Get a Job: Comparison of Survival Analysis of Semi-Parametric & Parametric Models

Abstract

Education is very important along with the advancements in technology that required people to have in-depth and quality knowledge. A person who can get a job usually has a good educational background. How long it takes a person to get a job is affected by many factors. In statistical sciences, factors affecting how long a person gets a job after graduating from college can be studied using survival analysis. Survival analysis is a regression method used to examine the factors that influence the occurrence of one or more events. Several survival models can be used to estimate the impact of survival factors on how long a person gets a job, including the cox proportional hazard model and the accelerated failure time model. The cox proportional hazard model belongs to the semi-parametric approach where the hazard function essentially follows the non-parametric model while the independent variables follow the parametric model. The accelerated failure time model is a parametric model based on data distribution that can predict the time of the event in observed data. This study was conducted by comparing the cox proportional hazard model and the accelerated failure time model with the logistics distribution to determine if the conditions or data types are better identified with the cox proportional hazard model or the failure time model. time. The results obtained by considering the minimum AIC value, the accelerated failure time model has a better logistics distribution than the cox proportional hazard model.