An empirical sensitivity study of mixture proportion estimators
- 1 January 1976
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics - Theory and Methods
- Vol. 5 (12) , 1115-1125
- https://doi.org/10.1080/03610927608827428
Abstract
The sensitivity of several proposed estimators of the mixture proportions a = (a1,…,am)T defining the normal mixture density are investigated when the component densities pk are subjected to changes in location. The particular deviations studied are motivated by an application of this model to crop acreage assessment using satellite multispectral sensor data.Keywords
This publication has 5 references indexed in Scilit:
- PATTERN CLUSTERING BY MULTIVARIATE MIXTURE ANALYSISMultivariate Behavioral Research, 1970
- Estimating the components of a mixture of normal distributionsBiometrika, 1969
- Estimation of Parameters for a Mixture of Normal DistributionsTechnometrics, 1966
- On the Estimation of Relative Potency in Dilution (-Direct) Assays by Distribution-Free MethodsPublished by JSTOR ,1963
- On a Modification of Certain Rank TestsThe Annals of Mathematical Statistics, 1963