This package provides tools for creating agents with persistent state using R6 classes <https://cran.r-project.org/package=R6> and the ellmer package <https://cran.r-project.org/package=ellmer>. Tracks prompts, messages, and agent metadata for reproducible, multi-turn large language model sessions.

Encodes several methods for performing Mendelian randomization analyses with summarized data. Similar to the MendelianRandomization package, but with fewer bells and whistles, and less frequent updates. As described in Yavorska (2017) <doi:10.1093/ije/dyx034> and Broadbent (2020) <doi:10.12688/wellcomeopenres.16374.2>.

Large collection of multilabel datasets along with the functions needed to export them to several formats, to make partitions, and to obtain bibliographic information.

User-friendly Shiny apps for designing and evaluating phase I cancer clinical trials, with the aim to estimate the maximum tolerated dose (MTD) of a novel drug, using a Bayesian decision procedure based on logistic regression.

Summarize multiple biomarker responses of aquatic organisms to contaminants using Cliffâ s delta, as described in Pham & Sokolova (2023) <doi:10.1002/ieam.4676>.

According to a phenomenon known as "the wisdom of the crowds," combining point estimates from multiple judges often provides a more accurate aggregate estimate than using a point estimate from a single judge. However, if the judges use shared information in their estimates, the simple average will over-emphasize this common component at the expense of the judgesâ private information. Asa Palley & Ville Satopää (2021) "Boosting the Wisdom of Crowds Within a Single Judgment Problem: Selective Averaging Based on Peer Predictions" <https://papers.ssrn.com/sol3/Papers.cfm?abstract_id=3504286> proposes a procedure for calculating a weighted average of the judgesâ individual estimates such that resulting aggregate estimate appropriately combines the judges collective information within a single estimation problem. The authors use both simulation and data from six experimental studies to illustrate that the weighting procedure outperforms existing averaging-like methods, such as the equally weighted average, trimmed average, and median. This aggregate estimate -- know as "the knowledge-weighted estimate" -- inputs a) judges estimates of a continuous outcome (E) and b) predictions of others average estimate of this outcome (P). In this R-package, the function knowledge_weighted_estimate(E,P) implements the knowledge-weighted estimate. Its use is illustrated with a simple stylized example and on real-world experimental data.

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.

This package provides samplers for various matrix variate distributions: Wishart, inverse-Wishart, normal, t, inverted-t, Beta type I, Beta type II, Gamma, confluent hypergeometric. Allows to simulate the noncentral Wishart distribution without the integer restriction on the degrees of freedom.

Analyse and visualise multi electrode array data at the single electrode and whole well level, downstream of AxIS Navigator 3.6.2 Software processing. Compare bursting parameters between time intervals and recordings using the bar chart visualisation functions. Compatible with 12- and 24- well plates.

There is an increasing interest in investigating how the compositions of microbial communities are associated with human health and disease. In this package, we present a novel global testing method called aMiSPU, that is highly adaptive and thus high powered across various scenarios, alleviating the issue with the choice of a phylogenetic distance. Our simulations and real data analysis demonstrated that aMiSPU test was often more powerful than several competing methods while correctly controlling type I error rates.

Bending non-positive-definite (symmetric) matrices to positive-definite, using weighted and unweighted methods. Jorjani, H., et al. (2003) <doi:10.3168/jds.S0022-0302(03)73646-7>. Schaeffer, L. R. (2014) <http://animalbiosciences.uoguelph.ca/~lrs/ELARES/PDforce.pdf>.

Nonparametric estimation and inference of a non-decreasing monotone hazard ratio from a right censored survival dataset. The estimator is based on a generalized Grenander typed estimator, and the inference procedure relies on direct plugin estimation of a first order derivative. More details please refer to the paper "Nonparametric inference under a monotone hazard ratio order" by Y. Wu and T. Westling (2023) <doi:10.1214/23-EJS2173>.

Extends the mlr3 ML framework with methods for spatial objects. Data storage and prediction are supported for packages terra', raster and stars'.

Dimension-reduction methods aim at defining a score that maximizes signal diversity. Three approaches, tree weight, maximum entropy weights, and maximum variance weights are provided. These methods are described in He and Fong (2019) <DOI:10.1002/sim.8212>.

This package provides functionality for estimating cross-sectional network structures representing partial correlations while accounting for missing data. Networks are estimated via neighborhood selection or regularization, with model selection guided by information criteria. Missing data can be handled primarily via multiple imputation or a maximum likelihood-based approach, as demonstrated by Nehler and Schultze (2025a) <doi:10.31234/osf.io/qpj35> and Nehler and Schultze (2025b) <doi:10.1080/00273171.2025.2503833>. Deletion-based approaches are also available but play a secondary role.

Estimation functions and diagnostic tools for mean length-based total mortality estimators based on Gedamke and Hoenig (2006) <doi:10.1577/T05-153.1>.

Friendly implementation of the Mann-Whitney-Wilcoxon test for competitive gene set enrichment analysis.

This package contains functions to access movement data stored in movebank.org as well as tools to visualize and statistically analyze animal movement data, among others functions to calculate dynamic Brownian Bridge Movement Models. Move helps addressing movement ecology questions.

Functions, data sets and examples for the book: Yves Croissant (2025) "Microeconometrics with R", Chapman and Hall/CRC The R Series <doi:10.1201/9781003100263>. The package includes a set of estimators for models used in microeconometrics, especially for count data and limited dependent variables. Test functions include score test, Hausman test, Vuong test, Sargan test and conditional moment test. A small subset of the data set used in the book is also included.

Causal moderated mediation analysis using the methods proposed by Qin and Wang (2023) <doi:10.3758/s13428-023-02095-4>. Causal moderated mediation analysis is crucial for investigating how, for whom, and where a treatment is effective by assessing the heterogeneity of mediation mechanism across individuals and contexts. This package enables researchers to estimate and test the conditional and moderated mediation effects, assess their sensitivity to unmeasured pre-treatment confounding, and visualize the results. The package is built based on the quasi-Bayesian Monte Carlo method, because it has relatively better performance at small sample sizes, and its running speed is the fastest. The package is applicable to a treatment of any scale, a binary or continuous mediator, a binary or continuous outcome, and one or more moderators of any scale.

Correct identification and handling of missing data is one of the most important steps in any analysis. To aid this process, mde provides a very easy to use yet robust framework to quickly get an idea of where the missing data lies and therefore find the most appropriate action to take. Graham WJ (2009) <doi:10.1146/annurev.psych.58.110405.085530>.

This package provides an extension to the lolog package by introducing the minTriadicClosure() statistic to capture higher-order interactions among triplets of nodes. This function facilitates improved modelling of group formations and triadic closure in networks. A smoothing parameter has been incorporated to avoid numerical errors.

Most multilevel methodologies can only model macro-micro multilevel situations in an unbiased way, wherein group-level predictors (e.g., city temperature) are used to predict an individual-level outcome variable (e.g., citizen personality). In contrast, this R package enables researchers to model micro-macro situations, wherein individual-level (micro) predictors (and other group-level predictors) are used to predict a group-level (macro) outcome variable in an unbiased way.

Rudimentary functions for sampling and calculating density from the matrix-variate variance-gamma distribution.