An implementation of the multilevel (also known as mixed or random effects) hidden Markov model using Bayesian estimation in R. The multilevel hidden Markov model (HMM) is a generalization of the well-known hidden Markov model, for the latter see Rabiner (1989) <doi:10.1109/5.18626>. The multilevel HMM is tailored to accommodate (intense) longitudinal data of multiple individuals simultaneously, see e.g., de Haan-Rietdijk et al. <doi:10.1080/00273171.2017.1370364>. Using a multilevel framework, we allow for heterogeneity in the model parameters (transition probability matrix and conditional distribution), while estimating one overall HMM. The model can be fitted on multivariate data with either a categorical, normal, or Poisson distribution, and include individual level covariates (allowing for e.g., group comparisons on model parameters). Parameters are estimated using Bayesian estimation utilizing the forward-backward recursion within a hybrid Metropolis within Gibbs sampler. Missing data (NA) in the dependent variables is accommodated assuming MAR. The package also includes various visualization options, a function to simulate data, and a function to obtain the most likely hidden state sequence for each individual using the Viterbi algorithm.

Implementation of the Marine Predators Algorithm (MPA) in R. MPA is a nature-inspired optimization algorithm that follows the rules governing optimal foraging strategy and encounter rate policy between predator and prey in marine ecosystems. Based on the paper by Faramarzi et al. (2020) <doi:10.1016/j.eswa.2020.113377>.

Fused lasso method to cluster and estimate regression coefficients of the same covariate across different data sets when a large number of independent data sets are combined. Package supports Gaussian, binomial, Poisson and Cox PH models.

Perform multi-trait rare-variant association tests using the summary statistics and adjust for possible sample overlap. Package is based on "Multi-Trait Analysis of Rare-Variant Association Summary Statistics using MTAR" by Luo, L., Shen, J., Zhang, H., Chhibber, A. Mehrotra, D.V., Tang, Z., 2019 (submitted).

Calculates Model-Averaged Tail Area Wald (MATA-Wald) confidence intervals, and MATA-Wald confidence densities and distributions, which are constructed using single-model frequentist estimators and model weights. See Turek and Fletcher (2012) <doi:10.1016/j.csda.2012.03.002> and Fletcher et al (2019) <doi:10.1007/s10651-019-00432-5> for details.

Several functions can be used to analyze multiblock multivariable data. If the input is a single matrix, then principal components analysis (PCA) is implemented. If the input is a list of matrices, then multiblock PCA is implemented. If the input is two matrices, for exploratory and objective variables, then partial least squares (PLS) analysis is implemented. If the input is two lists of matrices, for exploratory and objective variables, then multiblock PLS analysis is implemented. Additionally, if an extra outcome variable is specified, then a supervised version of the methods above is implemented. For each method, sparse modeling is also incorporated. Functions for selecting the number of components and regularized parameters are also provided.

The rapid screening of effective and optimal therapies from large numbers of candidate combinations, as well as exploring subgroup efficacy, remains challenging, which necessitates innovative, integrated, and efficient trial designs(Yuan, Y., et al. (2016) <doi:10.1002/sim.6971>). MIDAS-2 package enables quick and continuous screening of promising combination strategies and exploration of their subgroup effects within a unified platform design framework. We used a regression model to characterize the efficacy pattern in subgroups. Information borrowing was applied through Bayesian hierarchical model to improve trial efficiency considering the limited sample size in subgroups(Cunanan, K. M., et al. (2019) <doi:10.1177/1740774518812779>). MIDAS-2 provides an adaptive drug screening and subgroup exploring framework to accelerate immunotherapy development in an efficient, accurate, and integrated fashion(Wathen, J. K., & Thall, P. F. (2017) <doi: 10.1177/1740774517692302>).

This package provides functions and datasets used in the book: Fernandez-Casal, R., Costa, J. and Oviedo-de la Fuente, M. (2024) "Metodos predictivos de aprendizaje estadistico" <https://rubenfcasal.github.io/aprendizaje_estadistico/>.

Computes the optimal number of regions (or subdivisions) and their position in serial structures without a priori assumptions and to visualize the results. After reducing data dimensionality with the built-in function for data ordination, regions are fitted as segmented linear regressions along the serial structure. Every region boundary position and increasing number of regions are iteratively fitted and the best model (number of regions and boundary positions) is selected with an information criterion. This package expands on the previous regions package (Jones et al., Science 2018) with improved computation and more fitting and plotting options.

Analyzes production and dispersal of seeds dispersed from trees and recovered in seed traps. Motivated by long-term inventory plots where seed collections are used to infer seed production by each individual plant.

Implementation of various statistical models for multivariate event history data <doi:10.1007/s10985-013-9244-x>. Including multivariate cumulative incidence models <doi:10.1002/sim.6016>, and bivariate random effects probit models (Liability models) <doi:10.1016/j.csda.2015.01.014>. Modern methods for survival analysis, including regression modelling (Cox, Fine-Gray, Ghosh-Lin, Binomial regression) with fast computation of influence functions.

The Markov Decision Processes (MDP) toolbox proposes functions related to the resolution of discrete-time Markov Decision Processes: finite horizon, value iteration, policy iteration, linear programming algorithms with some variants and also proposes some functions related to Reinforcement Learning.

Monolix is a tool for running mixed effects model using saem'. This tool allows you to convert Monolix models to rxode2 (Wang, Hallow and James (2016) <doi:10.1002/psp4.12052>) using the form compatible with nlmixr2 (Fidler et al (2019) <doi:10.1002/psp4.12445>). If available, the rxode2 model will read in the Monolix data and compare the simulation for the population model individual model and residual model to immediately show how well the translation is performing. This saves the model development time for people who are creating an rxode2 model manually. Additionally, this package reads in all the information to allow simulation with uncertainty (that is the number of observations, the number of subjects, and the covariance matrix) with a rxode2 model. This is complementary to the babelmixr2 package that translates nlmixr2 models to Monolix and can convert the objects converted from monolix2rx to a full nlmixr2 fit. While not required, you can get/install the lixoftConnectors package in the Monolix installation, as described at the following url <https://monolixsuite.slp-software.com/r-functions/2024R1/installation-and-initialization>. When lixoftConnectors is available, Monolix can be used to load its model library instead manually setting up text files (which only works with old versions of Monolix').

The MARSS package provides maximum-likelihood parameter estimation for constrained and unconstrained linear multivariate autoregressive state-space (MARSS) models, including partially deterministic models. MARSS models are a class of dynamic linear model (DLM) and vector autoregressive model (VAR) model. Fitting available via Expectation-Maximization (EM), BFGS (using optim), and TMB (using the marssTMB companion package). Functions are provided for parametric and innovations bootstrapping, Kalman filtering and smoothing, model selection criteria including bootstrap AICb, confidences intervals via the Hessian approximation or bootstrapping, and all conditional residual types. See the user guide for examples of dynamic factor analysis, dynamic linear models, outlier and shock detection, and multivariate AR-p models. Online workshops (lectures, eBook, and computer labs) at <https://atsa-es.github.io/>.

This package provides a supervised learning algorithm inputs a train set, and outputs a prediction function, which can be used on a test set. If each data point belongs to a subset (such as geographic region, year, etc), then how do we know if subsets are similar enough so that we can get accurate predictions on one subset, after training on Other subsets? And how do we know if training on All subsets would improve prediction accuracy, relative to training on the Same subset? SOAK, Same/Other/All K-fold cross-validation, <doi:10.48550/arXiv.2410.08643> can be used to answer these questions, by fixing a test subset, training models on Same/Other/All subsets, and then comparing test error rates (Same versus Other and Same versus All). Also provides code for estimating how many train samples are required to get accurate predictions on a test set.

Process OpenPose human body keypoints for computer vision, including data structuring and user-defined linear transformations for standardization. It optionally, includes metadata extraction from filenames in the UCLA NewsScape archive.

This package provides a flexible general-purpose toolbox for implementing Rescorla-Wagner models in multi-armed bandit tasks. As the successor and functional extension of the binaryRL package, multiRL modularizes the Markov Decision Process (MDP) into six core components. This framework enables users to construct custom models via intuitive if-else syntax and define latent learning rules for agents. For parameter estimation, it provides both likelihood-based inference (MLE and MAP) and simulation-based inference (ABC and RNN), with full support for parallel processing across subjects. The workflow is highly standardized, featuring four main functions that strictly follow the four-step protocol (and ten rules) proposed by Wilson & Collins (2019) <doi:10.7554/eLife.49547>. Beyond the three built-in models (TD, RSTD, and Utility), users can easily derive new variants by declaring which variables are treated as free parameters.

This package provides functions for fitting various models to capture-recapture data including mixed-effects Cormack-Jolly-Seber(CJS) and multistate models and the multi-variate state model structure for survival estimation and POPAN structured Jolly-Seber models for abundance estimation. There are also Hidden Markov model (HMM) implementations of CJS and multistate models with and without state uncertainty and a simulation capability for HMM models.

Stand-alone HTTP capable R-package repository, that fully supports R's install.packages() and available.packages(). It also contains API endpoints for end-users to add/update packages. This package can supplement miniCRAN', which has functions for maintaining a local (partial) copy of CRAN'. Current version is bare-minimum without any access-control or much security.

Implementation of the sampling and aggregation method for the covariate shift maximin effect, which was proposed in <arXiv:2011.07568>. It constructs the confidence interval for any linear combination of the high-dimensional maximin effect.

This package provides a simple tool allowing users to easily and dynamically explore or document a data set using a tree structure.

Builds and interprets multi-response machine learning models using tidymodels syntax. Users can supply a tidy model, and mrIML automates the process of fitting multiple response models to multivariate data and applying interpretable machine learning techniques across them. For more details see Fountain-Jones (2021) <doi:10.1111/1755-0998.13495> and Fountain-Jones et al. (2024) <doi:10.22541/au.172676147.77148600/v1>.

Fits a Bayesian Regression Model for multivariate count data. This model assumes that the data is distributed according to the Conway-Maxwell-Poisson distribution, and for each response variable it is associate different covariates. This model allows to account for correlations between the counts by using latent effects based on the Chib and Winkelmann (2001) <http://www.jstor.org/stable/1392277> proposal.

This package provides a set of basic and extensible data structures and functions for multivariate analysis, including dimensionality reduction techniques, projection methods, and preprocessing functions. The aim of this package is to offer a flexible and user-friendly framework for multivariate analysis that can be easily extended for custom requirements and specific data analysis tasks.