Parameter Estimation for the Weibull Distribution under Interval Censored Test Plan with Random Blocks
Interval censoring is frequently encountered in industrial and biomedical data analysis.Previ-ous studies usually assumed that the data were independent identically distributed.However,in practical applications,the raw materials used by test units,test platforms and other factors can make the data have an obvious block structure,which leads to the correlation of data under the same block.In this paper,statistical analysis for the Weibull distribution model to the interval censored data with random block effects is considered.Three estimation methods,including two-stage,maximum likelihood and Bayes,are proposed to estimate the model parameters and predict the life times of the products.Meanwhile,a method based on Monte Carlo simulation is proposed to find the optimal test plan.The numerical results based on the simulated data and the real life example show that the proposed methods are feasible and effective.
random block effectinterval censoredWeibull distributionGibbs samplingGauss-Hermite approximationoptimal test plan
万方数据
维普
