Improved likelihood inferences for Weibull regression model

• Main Authors: SHEN, Yan, YANG, Zhenlin
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2017
• Subjects: Bias correction; variance correction; bootstrap; improved t-ratios; stochastic expansion; right censoring; Econometrics
• Online Access: https://ink.library.smu.edu.sg/soe_research/2080; https://ink.library.smu.edu.sg/context/soe_research/article/3080/viewcontent/ShenYang2016.pdf
• Institution: Singapore Management University

A general procedure is developed for bias-correcting the maximum likelihood estimators (MLEs) of the parameters of Weibull regression model with either complete or right-censored data. Following the bias correction, variance corrections and hence improved t-ratios for model parameters are presented. Potentially improved t-ratios for other reliability-related quantities are also discussed. Simulation results show that the proposed method is effective in correcting the bias of the MLEs, and the resulted t-ratios generally improve over the regular t-ratios.