Robustness of the Probability of Correct Selection Against Deviations from the Assumption of a Common Known Variance
| Autor: | P.C.T. van der Laan, B. van Putten, S.G.A.J. Driessen |
|---|---|
| Rok vydání: | 1990 |
| Předmět: |
Statistics and Probability
Unequal variances General Medicine Variance (accounting) Indifference zone approach Indifference zone Wiskundige en Statistische Methoden - Biometris Upper and lower bounds Subset selection approach Normal means selection procedure Robustness (computer science) Statistics Least favourable configuration Statistics Probability and Uncertainty Robustness Mathematical and Statistical Methods - Biometris Selection Selection (genetic algorithm) Mathematics |
| Zdroj: | Biometrical Journal, 32, 131-142 Biometrical Journal, 32(2), 131-142. Wiley-VCH Verlag Biometrical Journal 32 (1990) |
| ISSN: | 1521-4036 0323-3847 |
| DOI: | 10.1002/bimj.4710320202 |
| Popis: | In this paper we investigate the robustness of the probability of correct selection against departures from the assumption of a common known variance for the Indifference Zone approach (cf. BECHHOFER (1954)), as well as for the Subset Selection approach (cf. GUPTA (1956, 1965)) in the case of Normal populations. It appears that there is a serious lack of robustness in the sense that the actual lower bound of the probability of correct selection can be considerably smaller than the pretended lower bound P* of the probability of correct selection based on the assumption of a common known variance, if P* is taken 0.75, 0.90, 0.95 and 0.99, respectively. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Plný text