Randomization Effects in Hypothesis Testing with Autocorrelated Noise.
Abstract
In the presence of white noise, randomization is a powerful tool in the design of an experiment where the purpose of the experiment is to accept or reject one or more hypotheses. Randomization allows the experimentor to eliminate the influence of uncontrollable biases or trends upon the results of the experiment. In the presence of colored noise, however, the use of randomization may have a catastrophic effect upon the results of the hypothesis testing. The effects of several common randomization schemes upon the variance ratio test (F-test) are investigated under the assumption the noise is autocorrelated. Detailed results are presented by using a linear first-order Markov noise model in each randomization scheme. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 1976
- Accession Number
- ADA022891
Entities
People
- Elton P. Avara
Organizations
- United States Army Communications-Electronics Command