A uniform approximation to the sampling distribution of the coefficient of variation
According to Hendricks and Robey (1936) the coefficient of variation from a normal population with sample size n can be approximated by a function defined on the positive real line, which depends on the standard normal moment of order n - 1 about some well-defined point. Simple conditions under which this approximation is valid are derived. It is shown that the approximation error depends upon a standard normal stop-loss moment of order n - 1 about some point. As a main result we obtain a uniform error bound to the exact sampling density of the order of magnitude exp (- n/2k2), where k is the coefficient of variation.
Year of publication: |
1995
|
---|---|
Authors: | Hürlimann, Werner |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 24.1995, 3, p. 263-268
|
Publisher: |
Elsevier |
Keywords: | Coefficient of variation Sampling distribution Stop-loss error function |
Saved in:
Saved in favorites
Similar items by person
-
Lineare Programmierung : Eine Einführung für Kaufleute
Hürlimann, Werner, (1965)
-
Methodenkatalog : ein systematisches Inventar von über 3000 Problemlösungsmethoden
Hürlimann, Werner, (1981)
-
Simple risk forecasts using credibility
Hürlimann, Werner, (1988)
- More ...