From the 1 of 11 linked papers with an AI index.
11 papers · 1 filter
Handling mild outliers and unobserved values in compositional datasets using finite mixtures of mean-parametrised Dirichlet models
Jason Pillay, Andriëtte Bekker, Cristina Tortora +1
Heterogeneous compositional data may be simultaneously affected by missing values and atypical points, posing challenges for both clustering and outlier detection. We develop a mix…
A New Look at Gaussian Mixtures in the Presence of Missing-at-Random Responses and Covariates
Hung Tong, Antonio Punzo, Cristina Tortora
Missing values present a common challenge in statistical modeling, so handling them properly is an important research direction. Among the various mechanisms that can generate miss…
Handling Missingness and Censoring in Dirichlet Mixture Models
Jason Pillay, Andriette Bekker, Cristina Tortora +1
The paper introduces a likelihood‑based EM algorithm for fitting finite mixtures of Dirichlet distributions to compositional data that contain missing or left‑censored components,…
Handling Missingness and Censoring in Dirichlet Models
J. Pillay, A. Bekker, C. Tortora +1
Likelihood-based inference for compositional data generally requires fully observed compositions, hindering the direct treatment of missing or censored components on the simplex. I…
Modelling and detecting mild and gross anomalies in circular data via double-contaminated models
Antonio Punzo, Andriëtte Bekker, Arno Otto +2
In this paper, we propose a model-based framework to robustify inference for circular data in the presence of anomalous observations, distinguishing between mild and gross anomalie…
A Contaminated Model for Overdispersed Multinomial Microbiome Count Data
Ockert van Heerden, Andriëtte Bekker, Seite Makgai +2
Multinomial count data, such as microbial composition profiles derived from sequencing studies, frequently contain anomalous observations that distort parameter estimates. The Diri…