A New Range Statistic for Comparisons of Several Exponential Location Parameters
Abstract
Given k independent exponential populations with unknown and possibly unequal location parameters, with a common but unknown scale parameter, a complete random sample of common size n is drawn from each population. Based on these k samples, this paper proposes a new range statistic for comparing these exponential location parameters by a set of simultaneous confidence intervals. Upper percentage points are given, and the power function is also studied. Other applications in multiple range testing and in ranking and selection problems are noted. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 1981
- Accession Number
- ADA099399
Entities
People
- Hubert J. Chen
Organizations
- Ohio State University