answersLogoWhite

0

There are two broad cases: either you know the distribution or you don't.

Distribution known: Procedures are known for many, many situations. In some cases, it's possible to transform given data to fit available procedures. In other situations, it might be necessary to create new procedures.

Distribution unknown: Often a procedure that was developed for known distributions has been shown to work for distributions that are only similar to the known distributions. Recourse may alternatively be had to the so-called nonparametric statistics, that make minimal assumptions about distributions.

User Avatar

Wiki User

11y ago

What else can I help you with?