Robust estimation of a normal mixture
A robust procedure for estimating the five parameters of a mixture of two normals is proposed. The procedure is based on a robustification of the steps of the EM algorithm. This results in the use of an iteratively reweighted median and MAD. Simulation results are presented that show the superiority of the robustified EM when compared to the usual EM algorithm estimates.
Year of publication: |
1990
|
---|---|
Authors: | Veaux, Richard D. De ; Krieger, Abba M. |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 10.1990, 1, p. 1-7
|
Publisher: |
Elsevier |
Keywords: | Iteratively reweighted median iteratively reweighted MAD EM algorithm maximum likelihood estimation |
Saved in:
Online Resource
Saved in favorites
Similar items by person
-
Sharpe, Norean R., (2018)
-
Sharpe, Norean R., (2010)
-
Sharpe, Norean R., (2012)
- More ...