Non-parametric randomness tests based on success runs of fixed length

Let Xn be a random variable enumerating the number of appearances of a specific pattern in a sequence of n Bernoulli trials. A new method is presented for obtaining the conditional distribution of Xn given the number of successes in the n trials. The method is applied to three fixed-length run statistics and the results are used for establishing and investigating certain non-parametric tests of randomness.

MoreLess

Year of publication:	1997
Authors:	Koutras, M. V. ; Alexandrou, V. A.
Published in:	Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 32.1997, 4, p. 393-404
Publisher:	Elsevier
Keywords:	Bernoulli trials Success runs Non-parametric tests of randomness Markov chain imbeddable variables Generating functions

More details

Type of publication:	Article
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10005319803