Partitioning ordered variables into discrete states for discriminant analysis of ecological classifications

Publisher Website
Google Scholar
Abstract
This paper describes a procedure to partition ordered variables into discrete states for the discrimination of an ecological classification. At each step, the best partition is that which maximizes a log likelihood ratio for nonhomogeneity of distributions across the groups of the classification. The partitioning procedure ends when the probability of the log likelihood χ2 statistic reaches its minimum value. An actual ecological example is given of the discrete discriminant analysis of a benthos classification, with vegetation and oxygen concentration as discriminant variables. The O2 observations are first partitioned into discrete states, using the partitioning algorithm described before. Following three-dimensional contingency table analysis, it is concluded that the benthos classification is independent of oxygen concentrations.

This publication has 1 reference indexed in Scilit: