Interval estimation for parameters of Bathtub hazard model with fixed covariate in the presence of right and interval censored data

• Published in: AIP Conference Proceedings
• Main Author: Ismail I.; Arasan J.; Mustafa M.S.; Safari M.A.M.
• Format: Conference paper
• Language: English
• Published: American Institute of Physics 2024
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85188466925&doi=10.1063%2f5.0193257&partnerID=40&md5=cd93787b66fc316fa1bec63e967bc36b

A two-parameter bathtub hazard model was extended in this research. Its failure rate function can be increasing and bathtub-shaped depending on its parameters. The bathtub hazard model is an important lifetime model as it can accommodate both monotonic and non-monotonic failure rates. The model was extended by incorporating covariates in the presence of right-and interval-censored data. The parameters in the model were estimated using maximum likelihood estimation. Simulation study was done to evaluate the performance of parameter estimates via the bias, standard error (SE) and root mean square error (RMSE) values at different sample sizes and censoring proportion levels. The simulation results suggested that larger sample sizes and smaller censoring proportion yield better estimates. Then, two confidence interval estimation methods: Wald and likelihood ratio were constructed, and performance of these two methods were then assessed based on coverage probability study. The results suggested that Wald method works well when the sample size is greater than 30 and produces more asymmetrical intervals. Similar to the Wald method, likelihood ratio method appears to perform better when sample size is larger. © 2024 Author(s).