Applications of Ranked Set Sampling to Genetic Studies

• Activity Number: 93
• Type: Topic Contributed
• Date/Time: Monday, August 4, 2008 : 8:30 AM to 10:20 AM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Gang Zheng*+ and Chen Zehua and Kaushik Ghosh and Zhaohai Li
• Companies: National Heart, Lung, and Blood Institute and National University of Singapore and University of Nevada, Las Vegas and The George Washington University
• Keywords: ranked set sampling ; genetic association ; linkage ; cost-effective ; truncation ; simple random sampling

Abstract

In a typical application, ranked set sampling (RSS) is a cost-effective approach if ranking sampling units is easier and cheaper than measuring them. In this case, RSS is useful compared to classical random sampling. In genetic studies, genotypes and phenotypes (traits) of individuals have to be measured and obtained in genetic association and linkage studies. In many applications, genotyping cost is much more expensive than measuring phenotypes. Therefore, RSS can be applied. In this talk, we present two examples, where RSS can be easily applied, compared to using random sampling and the truncation approach. The first example is to map linkage disequilibrium and the second example is to test linkage using affected sib-pairs. In both examples, the idea of extreme rank selections will be presented. Simulation results will also be presented to compare different approaches.