Random KNN Classification and Regression (rknn)
Random knn classification and regression are implemented. Random knn based feature selection methods are also included. The approaches are mainly developed for high-dimensional data with small sample size.
Simultaneous Edit-Imputation for Continuous Microdata (EditImputeCont)
An integrated editing and imputation method for continuous microdata under linear constraints is implemented. It relies on a Bayesian nonparametric hierarchical modeling approach in which the joint distribution of the data is estimated by a flexible joint probability model. The generated edit-imputed data are guaranteed to satisfy all imposed edit rules, whose types include ratio edits, balance edits and range restriction
Vector Generalized Linear and Additive Models (VGAM)
An implementation of about 6 major classes of statistical regression models. At the heart of it are the vector generalized linear and additive model (VGLM/VGAM) classes. Currently only fixed-effects models are implemented, i.e., no random-effects models. Many (150+) models and distributions are estimated by maximum likelihood estimation (MLE) or penalized MLE, using Fisher scoring. VGLMs can be loosely thought of as multivariate GLMs. VGAMs are data-driven VGLMs (i.e., with smoothing). The other classes are RR-VGLMs (reduced-rank VGLMs), quadratic RR-VGLMs, reduced-rank VGAMs, RCIMs (row-column interaction models)—these classes perform constrained and unconstrained quadratic ordination (CQO/UQO) models in ecology, as well as constrained additive ordination (CAO). Note that these functions are subject to change, especially before version 1.0.0 is released; see the NEWS file for latest changes.