dc.contributor.author | Patel, Nitin R. | |
dc.contributor.author | Mehta, C. R. | |
dc.date.accessioned | 2010-09-20T04:59:08Z | |
dc.date.available | 2010-09-20T04:59:08Z | |
dc.date.copyright | 1980 | |
dc.date.issued | 1980-09-20T04:59:08Z | |
dc.identifier.citation | Communications in Statistics, 1980, 89, pp. 649-664 | en |
dc.identifier.uri | http://hdl.handle.net/11718/8809 | |
dc.description.abstract | A common statistical problem encountered in biomedical
research is to test the hypothesis that the parameters of k
binomial populations are all equal. An exact test of significance
of this hypothesis is possible in principle, the appropriate null
distribution being a normalized product of k binomial coefficients.
However, the problem of computing the tail area of this
distribution can be formidable since it requires the enumeration
of all sets of k binomial coefficients whose product is less than
a given constant. Existing algorithms, all of which rely on
explicit enumeration to generate feasible binomial coefficients soon become computationally infeasible. In this paper, we develop
a novel technique which drastically reduces the computational
effort needed to obtain the exact P-value. The problem is transformed
into one of identifying feasible paths through a directed
acyclic network by means of backwards induction and implicit
enumeration. The technique can be generalized for the exact
treatment of the RxC contingency table. | |
dc.language.iso | en | en |
dc.title | A network algorithm for the exact treatment of the 2xk contingency table | en |
dc.type | Article | en |