Supplementary materials for

Discriminant function analysis in marine ecology: some oversights and their solutions

J. Wilson White, Benjamin I. Ruttenberg

Dept. Ecology, Evolution, and Marine Biology
University of California
Santa Barbara, CA 93106

Mar Ecol Prog Ser 329:301-305


This program uses randomization to generate a P-value for the discriminant function
jackknife reclassification success statistic.

Matlab m-file: discriminate.m
Excel workbook for data import: Data_Import.xls
User guide: DiscriminateUserGuide.pdf

EM Algorithm

This is a simple Matlab version of Millar's (1987) Expectation-Maximization algorithm for
determining the maximum-likelihood mixing proportions and group memberships of unknown
data using a training dataset. 

This code does not have an Excel input worksheet and requires slightly more experience with Matlab.
Instructions can be found in the m-file's annotated text.

See Millar's original paper for details:
Millar RB (1987) Maximum likelihood estimation of mixed stock fishery composition. Can J Fish Aquat Sci 44:583-590

Matlab m-file: EM.m