Abstract:
|
Tolerance intervals provide limits that are expected to contain at least a specified proportion of the sampled population at a given confidence level. Tolerance intervals have been used in many diverse applications in broad areas like ecology, laboratory medicine, and quality control. They are also commonly employed in government regulatory documents. Over the past decade, theoretical developments and numerical routines have advanced the capability to construct (approximate) tolerance intervals for more complex data problems. In this talk, we develop approximate pointwise tolerance intervals for semiparametric regression models. We discuss theoretical results that depend on the central limit theorem for smoothers. Simulation results are presented to demonstrate the coverage properties of this approach. The approach is then applied to an engineering dataset in an effort to construct statistically-based design limits.
|