On Tests for Uniformity: Neyman's Statistic and Statistics Based on Gaps and Stretches.
Solomon,Herbert, Stephens,Michael A
Abstract
Solomon,Herbert, Stephens,Michael A
Abstract
Distributions of two statistics arising in sampling from a uniform distribution are investigated. They are Neyman's smooth goodness-of-fit statistic of order two and the maximum of a gap or stretch between p points in a sample of size n, p less than or equal to n. Pearson curve fits are applied to each situation thereby extending percentile tables provided by F.N. David (1939) for Neyman's statistic of order two, and indicating usage in computing percentiles for the stretch statistic. The Pearson curve values are excellent approximations along the range of values of each statistic. (Author)
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Distributions of two statistics arising in sampling from a uniform distribution are investigated. They are Neyman's smooth goodness-of-fit statistic of order two and the maximum of a gap or stretch between p points in a sample of size n, p less than or equal to n. Pearson curve fits are applied to each situation thereby extending percentile tables provided by F.N. David (1939) for Neyman's statistic of order two, and indicating usage in computing percentiles for the stretch statistic. The Pearson curve values are excellent approximations along the range of values of each statistic. (Author)
Key concepts: Statistic, Statistics, Mathematics, Percentile, PRESS statistic, Order statistic, Goodness of fit, Ancillary statistic