Testing for dependency of Bernoulli variables
2007 (English)In: International Journal of Statistical Sciences, ISSN 1683-5603, Vol. 5Article in journal (Refereed) Published
The aim of this paper is to derive test procedures for studies where data consist of pairs of Bernoulli variables. Applications exist in, for example, ophthalmology and studies on matched pairs. Score tests and likelihood ratio tests are derived for testing the dependency between the Bernoulli variables. Multinomial logit models are used to incorporate explanatory variables. Test statistics for two particular models are thoroughly outlined. Numerical illustrations of these test statistics are presented in three examples, including one with visual impairment data.
Place, publisher, year, edition, pages
2007. Vol. 5
Dependent binary data, generalized linear models
Probability Theory and Statistics
IdentifiersURN: urn:nbn:se:su:diva-10188OAI: oai:DiVA.org:su-10188DiVA: diva2:176707