Monitoring Test for One-Sided Multiple Quality Indices

• Activity Number: 57 - Some Recent Developments and Applications of the Empirical Likelihood Method
• Type: Topic Contributed
• Date/Time: Sunday, July 30, 2017 : 4:00 PM to 5:50 PM
• Sponsor: International Chinese Statistical Association
• Author(s): Guangyu Zhu* and JIahua Chen
• Companies: University of British Columbia and University of British Columbia
• Keywords: Bootstrap ; Composite likelihood ; Density ratio model ; Empirical likelihood ; Multiple sample ; Random effect

Abstract

The reliability of a wood structure depends on the strength of lumber product from the market. Closely monitoring the dynamic wood strength distribution of the solid lumber over the time is therefore an important quality control task. Due to the potential catastrophic event could be induced by a few weak wood piece, the lower quantiles of the strength distribution received most attention in the subject discipline. Clearly, monitoring only a single lower quantile of the wood strength is not sufficient. Even if the strength distribution of the wood product meets the quality standard specified on lower quantiles, the median or mean strengths could be significantly lower than the norm. In this presentation, we combine empirical likelihood and re-sampling method to construct a monitoring test for multiple quality indices. The proposed test tightly controls the type I error and is sensitive at detecting the quality reduction.