Summary
This paper presents two methods for modifying a statistical estimation or model fitting procedure. The first shows how to extend the procedure to mixture distributions. The second shows how to make the procedure more robust using weighting. The two modifications are then combined to produce robust methods for mixture analysis.
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
DEMPSTER, A.P., LAIRD, N.M., and RUBIN, D.B. (1977): Maximum-likeli– hood from Incomplete Data Via the EM Algorithm. Journal of the Royal Statistical Society, Series B, 39, pp. 1–38.
WINDHAM, M.P., (1994a) Mixture Analysis with Noisy Data. In New Approaches in Classification and Data Analysis. Proceedings of the Fourth Conference of the International Federation of Classification Societies (IFCS-93). E. Diday, Y. Lechavllier, M. Schader, P. Bertrand, B. Burtschy (eds.) Heidelberg: Springer-Verlag, 155–160.
WINDHAM, M.P., (1994b) Robustizing Model Fitting. Journal of the Royal Statistical Society, submitted.
WINDHAM M.P. and CUTLER A. (1992): Information Ratios for Validating Mixture Analyses. Journal of the American Statistical Association, 87, 1188–1192.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin · Heidelberg
About this paper
Cite this paper
Windham, M.P. (1996). Robustizing Mixture Analysis Using Model Weighting. In: Gaul, W., Pfeifer, D. (eds) From Data to Knowledge. Studies in Classification, Data Analysis, and Knowledge Organization. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-79999-0_10
Download citation
DOI: https://doi.org/10.1007/978-3-642-79999-0_10
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-60354-2
Online ISBN: 978-3-642-79999-0
eBook Packages: Springer Book Archive