# Mixtures of Contaminated Matrix Variate Normal Distributions

arXiv: https://arxiv.org/abs/2005.03861 · HTML: https://mlsanity.com/paper/2005.03861/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 May 2020.

## Built on by (most cited first, 2 of 2)

- [Mixtures of spatial factor analyzers for tensor-variate data](https://mlsanity.com/paper/2607.07887/) (2026)
- [Clustering Three-Way Data with Outliers](https://mlsanity.com/paper/2310.05288/) (2023)

## Abstract (arXiv)

Analysis of three-way data is becoming ever more prevalent in the literature, especially in the area of clustering and classification. Real data, including real three-way data, are often contaminated by potential outlying observations. Their detection, as well as the development of robust models insensitive to their presence, is particularly important for this type of data because of the practical issues concerning their effective visualization. Herein, the contaminated matrix variate normal distribution is discussed and then utilized in the mixture model paradigm for clustering. One key advantage of the proposed model is the ability to automatically detect potential outlying matrices by computing their textita posteriori probability to be a "good" or "bad" point. Such detection is currently unavailable using existing matrix variate methods. An expectation conditional maximization algorithm is used for parameter estimation, and both simulated and real data are used for illustration.