Non-centrality parameter

Some densities of non-central t-distributions

Non-centrality parameter (in short: NAP , English parameter noncentrality shortly NCP ) are parameters of families of probability distributions , the "central" in relation to other families of distributions are. While the central distribution describes how a test statistic is distributed if the null hypothesis is true, non-central distributions describe the distribution of a test statistic if the alternative hypothesis is true (i.e. the null hypothesis is false). This leads to the most important application of the parameters: the calculation of the selectivity . If the non-centrality parameter of a distribution is zero, then the distribution is identical to the distribution of the central family of distributions.

Use in probability distributions

Non-centrality parameters are used in the following distributions, among others:

• Non-central t-distribution
• Non-central chi-square distribution
• Non-central F distribution

Individual evidence

1. ^ Dodge, Y. (2003). The Oxford Dictionary of Statistical Terms , Oxford University Press. ISBN 0-19-920613-9