Robustizing Mixture Analysis Using Model Weighting
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.
KeywordsMixture Distribution Model Family Mixture Analysis Normal Mixture Asymptotic Efficiency
Unable to display preview. Download preview PDF.
- 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.Google Scholar
- 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.Google Scholar
- WINDHAM, M.P., (1994b) Robustizing Model Fitting. Journal of the Royal Statistical Society, submitted.Google Scholar