Tests for Uniformity Arising from a Series of Events.
Abstract
Test procedures are reviewed for testing whether events in time are occurring randomly. When this is so, a well-known transformation takes the times of occurrence to uniform order statistics, and the problem reduces to finding tests for uniformity which are sensitive to those alternative patterns of events which might be met in practice. Some alternatives are described and use is discussed of Neyman's test procedure, EDF tests, and spacings. Some recent work on spacings is examined in this context, and throughout the paper suggestions are made for future work. Keywords: Exponential distribution; Renewal processes; Gamma distribution; Weibull distribution; Uniform distribution; Tests for exponentiality; EDF tests; Neyman's tests; Spacings.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 28, 1988
- Accession Number
- ADA189069
Entities
People
- Michael A. Stephens
Organizations
- Stanford University