A folded model for compositional data analysis
This is an incremental improvement for researchers in compositional data analysis, offering a more flexible distribution class for simplex-based data.
The authors tackled the problem of analyzing compositional data by developing a folded model that extends the α-transformation, resulting in better performance in capturing data structure compared to the logistic normal distribution and a similar non-folded model.
A folded type model is developed for analyzing compositional data. The proposed model involves an extension of the $α$-transformation for compositional data and provides a new and flexible class of distributions for modeling data defined on the simplex sample space. Despite its rather seemingly complex structure, employment of the EM algorithm guarantees efficient parameter estimation. The model is validated through simulation studies and examples which illustrate that the proposed model performs better in terms of capturing the data structure, when compared to the popular logistic normal distribution, and can be advantageous over a similar model without folding.