https://lhzmix.r-universe.devRecent package updates in lhzmixR-universehttps://github.com/lhzmix.pnghttps://lhzmix.r-universe.devMon, 07 Sep 2026 23:48:44 GMThanzhang.lu@ubc.ca (Hanzhang Lu)Fits single-component and finite-mixture Kronecker-structured matrix-normal models and imputes incomplete matrix-variate data using matrix partial expectation-maximization. General MPEM handles arbitrary missingness, while Rect-MPEM exploits rectangular structural missingness. The methods are described in Lu, Andrews and Browne (2026) "An Efficient EM Algorithm for Both Element-Wise and Structural Missingness in Matrix-Variate Normal Mixture Models" <doi:10.48550/arXiv.2609.00616>.https://github.com/r-universe/lhzmix/actions/runs/35440180180Mon, 07 Sep 2026 23:48:44 GMTmpem0.1.0successhttps://lhzmix.r-universe.devhttps://github.com/lhzmix/mpemhanzhang.lu@ubc.ca (Hanzhang Lu)Fits spatial mixture models, including spatial Gaussian mixtures and mixtures of spatial factor analyzers, to complete or incomplete data. Spatial decay can be represented by monotone I-splines or a normalized sigmoid. Missing entries are handled using a built-in partial expectation-maximization procedure for matrix-variate data. The spatial covariance and spatial factor analyzer models are described in Lu and colleagues (2026a) "Spatial Covariance Constraints for Gaussian Mixture Models" <doi:10.48550/arXiv.2601.07979> and Lu and colleagues (2026b) "Mixtures of spatial factor analyzers for tensor-variate data" <doi:10.48550/arXiv.2607.07887>.https://github.com/r-universe/lhzmix/actions/runs/38152093990Thu, 27 Aug 2026 23:21:32 GMTSpatMix0.1.0successhttps://lhzmix.r-universe.devhttps://github.com/lhzmix/spatmix