On the Behavior of Prior Distributions Fitted to Failure Data.
Abstract
Equipment mean time between failures (MTBF) is assumed to be a random variable, the probability density function of the MTBF being called the prior density. The method of moments procedure for fitting the uniform prior density to failure data is given. The goodness of fit of the uniform prior is compared to the goodness of fit of the inverted gamma prior for actual failure data. Posterior producer's and consumer's risks for the uniform and inverted gamma priors are compared in the context of Bayesian reliability demonstration tests. It is shown that the corresponding posterior risks for the two priors can be significantly different even when the two priors fit the data equally well.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1975
- Accession Number
- ADA020677
Entities
People
- C. P. Tsokos
- J. J. Higgins
Organizations
- University of South Florida