Robust fitting of mixtures of factor analyzers using the trimmed likelihood estimator. (English) Zbl 1362.62061
Summary: Mixtures of factor analyzers (MFAs) have been popularly used to cluster the high-dimensional data. However, the traditional estimation method is based on the normality assumptions of random terms and thus is sensitive to outliers. In this article, we introduce a robust estimation procedure of MFAs using the trimmed likelihood estimator. We use a simulation study and a real data application to demonstrate the robustness of the trimmed estimation procedure and compare it with the traditional normality-based maximum likelihood estimate.
MSC:
| 62F35 | Robustness and adaptive procedures (parametric inference) |
| 62H30 | Classification and discrimination; cluster analysis (statistical aspects) |
| 62F10 | Point estimation |
Keywords:
EM algorithm; factor analysis; mixture models; robustness; trimmed likelihood estimator; mixtures of factor analyzers (MFAs)References:
| [1] | DOI: 10.1016/j.csda.2010.05.019 · Zbl 1247.62151 · doi:10.1016/j.csda.2010.05.019 |
| [2] | Arminger G., Psychometrika 65 pp 475– |
| [3] | DOI: 10.1093/bioinformatics/btr112 · doi:10.1093/bioinformatics/btr112 |
| [4] | Bishop C. M., Learning in Graphical Models pp 371– |
| [5] | Clark B. R., Probability and Statistics 30 pp 203– |
| [6] | DOI: 10.1080/01621459.2000.10474285 · doi:10.1080/01621459.2000.10474285 |
| [7] | DOI: 10.1016/S0167-7152(96)00145-9 · Zbl 1064.62540 · doi:10.1016/S0167-7152(96)00145-9 |
| [8] | DOI: 10.1007/BF02294853 · Zbl 1291.62201 · doi:10.1007/BF02294853 |
| [9] | DOI: 10.1023/A:1020297828025 · Zbl 1033.68085 · doi:10.1023/A:1020297828025 |
| [10] | Ghahramani Z., The EM Algorithm for Mixture of Factor Analyzers |
| [11] | DOI: 10.1016/j.jmva.2008.09.006 · Zbl 1157.62040 · doi:10.1016/j.jmva.2008.09.006 |
| [12] | DOI: 10.1214/009053604000000571 · Zbl 1047.62063 · doi:10.1214/009053604000000571 |
| [13] | DOI: 10.1109/72.554192 · doi:10.1109/72.554192 |
| [14] | DOI: 10.1214/088342305000000016 · Zbl 1100.62032 · doi:10.1214/088342305000000016 |
| [15] | DOI: 10.1007/BF00773479 · Zbl 0822.62019 · doi:10.1007/BF00773479 |
| [16] | DOI: 10.1016/j.csda.2006.09.015 · Zbl 1445.62053 · doi:10.1016/j.csda.2006.09.015 |
| [17] | DOI: 10.1002/0471721182 · doi:10.1002/0471721182 |
| [18] | DOI: 10.1016/S0167-9473(02)00183-4 · Zbl 1256.62036 · doi:10.1016/S0167-9473(02)00183-4 |
| [19] | DOI: 10.1016/j.csda.2006.12.024 · Zbl 1328.62033 · doi:10.1016/j.csda.2006.12.024 |
| [20] | Neykov N. M., Developments in Robust Statistics pp 277– |
| [21] | DOI: 10.1111/1467-9868.00265 · Zbl 0957.62020 · doi:10.1111/1467-9868.00265 |
| [22] | Tipping M. E., Mixtures of Probabilistic Principal Component Analysers |
| [23] | DOI: 10.1162/089976699300016728 · doi:10.1162/089976699300016728 |
| [24] | Vandev D. L., New Directions in Data Analysis and Robustness pp 257– |
| [25] | DOI: 10.1007/s11222-010-9226-8 · Zbl 1322.62047 · doi:10.1007/s11222-010-9226-8 |
| [26] | DOI: 10.1080/03610926.2010.526741 · Zbl 1244.62038 · doi:10.1080/03610926.2010.526741 |
| [27] | DOI: 10.1198/jasa.2009.0237 · Zbl 1388.62007 · doi:10.1198/jasa.2009.0237 |
| [28] | DOI: 10.1016/j.csda.2013.07.019 · Zbl 1471.62227 · doi:10.1016/j.csda.2013.07.019 |
| [29] | DOI: 10.1007/BF02294554 · Zbl 0890.62047 · doi:10.1007/BF02294554 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
