|
| << vorige | volgende >> |
Tijdschrift beschrijving
Alle jaargangen van het bijbehorende tijdschrift
Alle afleveringen van het bijbehorende jaargang
Alle artikelen van de bijbehorende aflevering
Details van artikel 16 van 19 gevonden artikelen
| Titel: |
Interquartile tests for dispersion in skewed distributions |
| Auteur: |
Shoemaker, Lewis H. |
| Verschenen in: |
Communications in statistics |
| Paginering: | Jaargang 28 (1999) nr. 1 pagina's 189-205 |
| Jaar: |
1999 |
| Inhoud: |
The estimation and testing of variability in data is an important problem in almost all areas of business and science, especially in the area of quality control. Most parametric and nonparametric tests for differences in scale require that the underlying distributions be approximately symmetric when location is unknown. In fact choice of location estimator will influence the distribution of the test statistics. The common two sample F-test for difference in variances is unreliable in all situations except when the population distributions are normal. Nonparametric tests are useful when the underlying distributions are heavy tailed, but most are unreliable when the distributions are skewed. The exception to this is when location is known. This paper introduces a class of tests for heterogeneous spreads which is valid for a variety of both skewed and symmetric distributions when location is known or unknown. In addition, the tests have intuitive estimators of spread associated with them. |
| Uitgever: |
Taylor & Francis |
| Bronbestand: |
Elektronische Wetenschappelijke Tijdschriften |
Details van artikel 16 van 19 gevonden artikelen
| << vorige | volgende >> |