PDF pending restoration
ERIC Number: ED476179
Record Type: Non-Journal
Publication Date: 2002-Apr
Pages: 192
Abstractor: N/A
ISBN: N/A
ISSN: N/A
EISSN: N/A
Comparison of K-Means Clustering with Linear Probability Model, Linear Discriminant Function, and Logistic Regression for Predicting Two-Group Membership.
So, Tak-Shing Harry; Peng, Chao-Ying Joanne
This study compared the accuracy of predicting two-group membership obtained from K-means clustering with those derived from linear probability modeling, linear discriminant function, and logistic regression under various data properties. Multivariate normally distributed populations were simulated based on combinations of population proportions, equality of covariance matrices, and group separation. The four statistical methods were applied to training samples drawn based on combinations of sample representativeness and sample size. Error rates were calculated based on the cross-validation results on test samples. The findings revealed that, depending on the data pattern, K-means clustering was a viable alternative when the accuracy of predicting the membership of the smaller population was the main objective. (Contains 25 figures, 32 tables, and 27 references.) (Author/SLD)
Descriptors: Cluster Grouping, Group Membership, Prediction, Probability, Research Methodology, Simulation
Publication Type: Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A
Grant or Contract Numbers: N/A