Bootstrap confidence intervals for the mode of log-logistic hazard function
By considering hazard function for log-logistic distribution with parameter Q > 1 , it is important to perform inferences about the mode of the hazard function with unimodal hazard function. The parameters of this distribution are estimated by maximum likelihood method and they are used to estima...
Saved in:
| 主要作者: | |
|---|---|
| 格式: | Thesis |
| 语言: | English |
| 出版: |
2009
|
| 主题: | |
| 在线阅读: | http://eprints.uthm.edu.my/4000/1/24p%20SITI%20NORMAH%20HASAN.pdf |
| 标签: |
添加标签
没有标签, 成为第一个标记此记录!
|
| 总结: | By considering hazard function for log-logistic distribution with parameter Q > 1 , it is important to perform inferences about the mode of the hazard function with unimodal hazard function. The parameters of this distribution are estimated by maximum likelihood method and they are used to estimate other quantities of interest such as mode of lifetime data and percentile. From the asymptotical normality of the maximum likelihood estimator, confidence intervals can be obtained. However, these results might not be very accurate for the small sample size or large proportion of censored data. In this project, the confidence interval for the mode of the hazard function obtained by asymptotic confidence interval is going to be compared with boatstrap methods. The performance of the procedures is evaluated by simulation with different sample sizes and proportion of censored data |
|---|