F distribution

(redirected from F-distribution)
Also found in: Dictionary, Medical, Wikipedia.
Related to F-distribution: T-distribution

F distribution

[′ef ‚dis·trə‚byü·shən]
(statistics)
The ratio of two independent chi-square variables each divided by its degree of freedom; used to test hypotheses in the analysis of variance and hypotheses about whether or not two normal populations have the same variance.
References in periodicals archive ?
Technically, we need to adjust the PLINK output p value using an F-distribution with different DFs (see supplementary materials for technical details).
In general, the proposed Wald tests follow the exact F-distribution under the null hypothesis and hence the type I errors are well controlled under all settings.
has a noncentral F-distribution with [v.sub.1] and [v.sub.2] degrees of freedom and noncentrality parameter [lambda].
has a doubly noncentral F-distribution with [v.sub.1] and [v.sub.2] degrees of freedom, and noncentrality parameters [[lambda].sub.1] and [[lambda].sub.2].
For any candidate h other than i, let B(|u.sub.i~) denote the probability that candidate h will get an approval vote from any given voter, when candidate i offers |u.sub.i~ and all other candidates' offers are drawn independently from the F-distribution.(5)
Candidate i would gain by deviating from the F-distribution to some other distribution G (when all other candidates are expected to use the F-distribution) if the deviation would cause candidate i's average approval score to become higher than the other candidates' average approval scores.
If the hypothesis that the four industries have the same elasticities is true, then the traditional test statistic involved has F-distribution with degrees of freedom 6 and 50.
If the hypothesis that the four food-processing industries have the same slope parameters ([[beta].sub.1] and [[beta].sub.3]) for their indirect forms (8) is true, then the appropriate test statistic has F-distribution with degrees of freedom 6 and 30, respectively.