Generally Usable Sequential Randomization Tests for Two-Way ANOVA That Emphasize the More Recent Data.
Abstract
The data are independent observations from two or more sources. Under the null hypothesis, observations from the same source have the same (arbitrary) distribution. Observations are obtained in successive groups each containing a set of specified size from each source (with a stated maximum total available from each source). An overall test is a succession of subtests with significance when at least one subtest is significant. All observations are re-used until a specified number of groups occur. Then, independently for each source, a stated number of the observations from a source are chosen by randomization (all possibilities equally likely). Data for re-use are now the group of observations chosen by the randomizations and newly obtained groups. Additional groups are taken until the number for re-use reaches a given value. Then, independently for each source, a stated number of the observations in these groups from a source are chosen by randomization. The resulting group is the data for re-use at this stage. Additional new groups are taken, etc. Exact null probabilities are obtainable, through use of appropriate randomization models and special kinds of subtest statistics. The overall test ends when a significant subtest occurs (thus saving time and expense). Some possible uses in quality control are outlined. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1970
- Accession Number
- AD0717931
Entities
People
- John E. Walsh
Organizations
- Southern Methodist University