Tolerance Intervals for One-Way Random Effects Models Based on Modified Signed Log-Likelihood Ratio

• Activity Number: 88
• Type: Contributed
• Date/Time: Sunday, August 1, 2010 : 4:00 PM to 5:50 PM
• Sponsor: Section on Quality and Productivity
• Author(s): Gaurav Sharma*+ and Thomas Mathew
• Companies: University of Maryland Baltimore County and University of Maryland Baltimore County
• Keywords: One-way random effects models ; Modified signed log-likelihood ratio ; Signed log-likelihood ratio ; Tolerance Intervals

Abstract

A tolerance interval, computed using a random sample, is an interval that will contain a specified proportion or more of a population, with a given confidence level. Tolerance intervals are widely used in many applications involving engineering, environmental and biological processes. This talk is on the construction of one-sided and two-sided tolerance intervals in a one-way random models under the usual assumptions, when the data are balanced or unbalanced. The construction of the tolerance intervals will be based on the concept of the modified signed log-likelihood ratio procedure (MSLRP). Simulation results and examples will be presented, along with a computational algorithm. It will be noted that the use of the MSLRP allows a straightforward computation of the tolerance intervals.