Search for: [Abstrakt = "We propose a variable selection for model\-based clustering using greedy search algorithm. At each stage it searches for the variable to add that most improves the clustering as measured by BIC. As a result the best combination of number of groups and clustering model is chosen \(two nested subsets of variables are recast as a model comparison problem\). We compare variable selection using greedy search algorithm with a HINoV method of variable selection. \(original abstract\)"]

Number of results: 1