| Autor: |
Beninel, Farid, Husson, François |
| Zdroj: |
Computational Statistics; Jul1999, Vol. 14 Issue 2, p251-261, 11p |
| Abstrakt: |
Summary: The study of the exact cumulative distribution function of discrete statistics, has often resulted in a strong dependence with regard to numbers of particular sums. For some statistics, the latter are derived from a finite sequence of real numbers (αi)1≤i≤N and have to be less than or equal to a given X value. Constraints over the cardinality of combinations can be proposed. When calculation are highly complex, approximated distributions are usually used, which are often inappropriate to small N values. In this paper, we provide a calculation technique which allows the exact distribution to be studied. Consequently, we can derive exact p-values associated with observed values of the considered statistics. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
Full text from SpringerLink