Study on the influence of prior reliability for bayesian estimation

Bayesian method has been widely used for parameters evaluation and decision making. However, existing studies paid more attention on the construction of prior distribution and method engineering applications. The previous researches had fewer discourses on the reliability influence analysis of prior information or prior distribution. In this paper, based on the coherence analysis between prior information and field test results, the definition of prior reliability was presented, and the calculation process based on hypothesis testing theory was given. For normal distribution, to a numerical example, the influence analysis result was exhibited. Furthermore, the interval estimation method considering the influence of prior reliability was given.

[1]  Min Xie,et al.  Bayes reliability demonstration test plan for series-systems with binomial subsystem data , 1998, Annual Reliability and Maintainability Symposium. 1998 Proceedings. International Symposium on Product Quality and Integrity.

[2]  John S. J. Hsu,et al.  Bayesian Methods: An Analysis for Statisticians and Interdisciplinary Researchers , 1999 .

[3]  Neil D. Lawrence,et al.  A variational approach to robust Bayesian interpolation , 2003, 2003 IEEE XIII Workshop on Neural Networks for Signal Processing (IEEE Cat. No.03TH8718).

[4]  Ronald V. Canfield.  Conservative Bayes experimental design , 1992 .

[5]  R. Schafer.  Bayesian Reliability Analysis , 1983 .

[6]  B. Bertsche,et al.  Analysis of the staircase method considering the Bayes theorem , 2005, Annual Reliability and Maintainability Symposium, 2005. Proceedings..

[7]  J. Berger.  Statistical Decision Theory and Bayesian Analysis , 1988 .