Abstract

Final project is an obligation of Mathematics Students at Udayana University in applying the theories learned during their studies. Ideally, the final project is expected to be completed no longer than 4 semester. In fact, there are several students who have not been able to accomplish this target due to many factors inhibit them. Therefore, this study was conducted to find out the model in describing the factors that affect the duration of the final project in mathematics students class of 2015-2019 with survival analysis, Bayesian Cox regression. This research data was obtained through distributing questionnaires using proportionate stratified random sampling. The data analysis stages are carried out by data collecting, pre-processing, descriptive analysis, outlier detection, proportional hazard test, Cox regression analysis with  Bayes approach, and forming a Bayesian Cox regression model from credible intervals. By using the Metropolis random walk algorithm in the MCMC simulation, normal priors on the regression coefficients, and based on credible intervals, the results obtained that there are 3 variables that have a significant influence, such as GPA value, getting motivation from friends/family, and living with family during the final project.