PLACE:  319 Snedecor

SPEAKER:  Yao Zhang, Department of Statistics, Iowa State University

TITLE: Bayesian Optimum Design for Accelerated Life Tests

ABSTRACT:

We present a Bayesian optimum design method for accelerated life tests with
one accelerating variable, when the acceleration model is linear in the
parameters, based on censored data from a log-location-scale distribution.
We develop a Bayes criterion based on the estimation precision of a
distribution quantile at a specified use condition and use this criterion
to find the optimum designs for the tests.  A large-sample normal
approximation provides an easy-to-interpret yet useful simplification to
this design problem. We present a numerical example using the Weibull
distribution with Type I censoring to illustrate the method and to examine
the prior, censoring, and sample size effects. The general equivalence
theorem for the proposed criterion is used to verify that the numerically
optimized designs are globally optimal. The resulting optimum designs are
evaluated through simulation.
 

COFFEE:  9:45 a.m., 104 Snedecor Hall