O'Reilly logo

Statistical Methods for Fuzzy Data by Reinhard Viertl

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

13.2 Fuzzy p-values

Let η(·) be the characterizing function of the fuzzy value t* = t(x1*, … , xn*) of a test statistic T = t(x1, … , xn) based on the fuzzy sample x1*, … , xn*.

In applications the support supp [η(·)] = {x : η(x) > 0} is usually a bounded subset of .

Considering the δ-cuts of t* for δ ∈ (0; 1] and denoting them by

Unnumbered Display Equation

the corresponding fuzzy p-value p* is defined via its δ-cuts Cδ(p*) for δ ∈ (0; 1] in the following way:

For a one-sided test with decision rule for Ttcr reject hypothesis, the δ-cuts of the corresponding fuzzy p-value p* are defined by

Unnumbered Display Equation

with t1(δ) and t2(δ) from above.

For one-sided tests with decision rule for Ttcr reject the hypothesis, the δ-cults of the corresponding fuzzy p-value p* are defined by

Unnumbered Display Equation

In the case of two-sided tests first it has to be decided on which side of the median m of the distribution of T the main part of the amount of fuzziness of t* is located. Therefore one has to compute the area under the characterizing function η(·) of ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required