You are not currently logged in.
Access JSTOR through your library or other institution:
If You Use a Screen ReaderThis content is available through Read Online (Free) program, which relies on page scans. Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
The Notion of Redundancy and Its Use as a Quantitative Measure of the Discrepancy between a Statistical Hypothesis and a Set of Observational Data [with Discussion]
Scandinavian Journal of Statistics
Vol. 1, No. 1 (1974), pp. 3-18
Published by: Wiley on behalf of Board of the Foundation of the Scandinavian Journal of Statistics
Stable URL: http://www.jstor.org/stable/4615544
Page Count: 16
You can always find the topics here!Topics: Statistics, Statistical models, Statism, Null hypothesis, Entropy, Data models, Degrees of freedom, Approximation, Children, Traffic accidents
Were these topics helpful?See somethings inaccurate? Let us know!
Select the topics that are inaccurate.
Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
Preview not available
It is proposed to supplement the critical level, as used in ordinary significance testing, by a measure of the magnitude of the departure of the observed set of data from the hypothesis to be tested. This measure, which is called the redundancy, appears in two versions, one microcanonical (or combinatorial) and the other canonical (or parametrical). The microcanomical redundancy is obtained by dividing minus the logarithm of the critical level by the Boltzmann entropy of the experiment and the canonical redundancy by dividing minus the logarithm of the likelihood ratio by the Gibbsian entropy. An approximation theorem shows that the former may be approximated asymptotically by the latter. The problem of calibrating the redundancy scale is discussed in connection with a series of examples, and, finally, certain considerations concerning the size of a statistical experiment are given which are based on the redundancy rather than the power function.
Scandinavian Journal of Statistics © 1974 Board of the Foundation of the Scandinavian Journal of Statistics