Variational Bayes Approximations for Clustering via Mixtures of Normal Inverse Gaussian Distributions
This work addresses clustering problems for data analysts by providing a more efficient method, though it is incremental as it adapts existing variational techniques to a specific distribution type.
The paper tackled parameter estimation for model-based clustering using finite mixtures of normal inverse Gaussian distributions by employing variational Bayes approximations, which reduced computational complexities compared to traditional EM methods, as demonstrated on simulated and real data.
Parameter estimation for model-based clustering using a finite mixture of normal inverse Gaussian (NIG) distributions is achieved through variational Bayes approximations. Univariate NIG mixtures and multivariate NIG mixtures are considered. The use of variational Bayes approximations here is a substantial departure from the traditional EM approach and alleviates some of the associated computational complexities and uncertainties. Our variational algorithm is applied to simulated and real data. The paper concludes with discussion and suggestions for future work.