distribution-free method

distribution-free method

[‚dis·trə′byü·shən ‚frē ‚meth·əd]
(statistics)
Any method of inference that does not depend on the characteristics of the population from which the samples are obtained.
References in periodicals archive ?
The aim of this paper is to present a distribution-free method based on an application of bootstrap method to obtain confidence intervals of Morisita's index, in a temporal sequence, in order to characterize the spatial-temporal dynamics of the coffee-leaf-miner (L.
The approach of smooth ROC curve fitting can be viewed as a distribution-free method; that is, it would perform consistently, irrespective of the true underlying distribution.
For many applications, rather than trying to find an appropriate statistical distribution or transformation to model the observed data, it would probably be better to abandon this classic approach and switch to modern distribution-free methods. For example, a popular approach that can avoid many of these problems is the generalized estimating equations, or GEE.
Like previous revisions since 1989, this edition reflects current levels of computing power and the scope of applying nonparametric or distribution-free methods, especially the tendency to combine them with other methods.
Nonparametric and distribution-free methods for the social sciences.
Full browser ?