A comparison of the mixture and classification approaches to cluster analysis
Authors:
- 1 January 1980
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics - Theory and Methods
- Vol. 9 (9) , 923-933
Abstract
This paper examines the relative performance of two commonly used clustering methods based on maximum likelihood in the context of classifying a sample of observations of unknown origin arising from two normal populations with a common covariance matrix. the associated properties of the two methods are compared by conducting a series of simulation experiments under both mixture and separate sampling schemes.Keywords
This publication has 20 references indexed in Scilit:
- Small sample results for a linear discriminant function estimated from a mixture of normal populationsJournal of Statistical Computation and Simulation, 1979
- A Case Study of two Clustering Methods based on Maximum LikelihoodStatistica Neerlandica, 1979
- Unresolved Problems in Cluster AnalysisBiometrics, 1979
- The efficiency of a linear discriminant function based on unclassified initial samplesBiometrika, 1978
- Asymptotic behaviour of classification maximum likelihood estimatesBiometrika, 1978
- Separate sample logistic discriminationBiometrika, 1972
- Estimating the components of a mixture of normal distributionsBiometrika, 1969
- Percentage Points of a Test for ClustersJournal of the American Statistical Association, 1969
- On Some Invariant Criteria for Grouping DataJournal of the American Statistical Association, 1967
- A Note on the Generation of Random Normal DeviatesThe Annals of Mathematical Statistics, 1958