Bayesian Methods for Skew-Normal Accelerated Destructive Degradation Test

• Main Authors: I-Ju Chen, 陳奕汝
• Other Authors: Chih-Chun Tsai
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/s255hr

碩士 === 淡江大學 === 數學學系數學與數據科學碩士班 === 106 === The accelerated destructive degradation test (ADDT) provided an effective way to assess the reliability information of the highly reliable products whose quality characteristics degraded over time, and can be taken only once on each tested unit during the measurement process. Motivated by a polymer data, Lin (2013) proposed a nonlinear ADDT model with measurement error that follows a skew-normal distribution, and derived the analytical expressions for the product''s lifetime distribution. However, the 95% confidence interval of the product''s 100pth percentile is not precision. Hence we used Bayesian approach improve the provision of the estimation. More specifically speaking, this article used Metropolis-Hasting algorithm to estimate the parameters of the model, and obtain the posterior credible interval. Finally, a simulation study was conducted to compare the precision of the maximum likelihood and Bayesian estimations.