An R command interface to the MLwiN multilevel modelling software package.

Rcpp bindings for PLANC', a highly parallel and extensible NMF/NTF (Non-negative Matrix/Tensor Factorization) library. Wraps algorithms described in Kannan et. al (2018) <doi:10.1109/TKDE.2017.2767592> and Eswar et. al (2021) <doi:10.1145/3432185>. Implements algorithms described in Welch et al. (2019) <doi:10.1016/j.cell.2019.05.006>, Gao et al. (2021) <doi:10.1038/s41587-021-00867-x>, and Kriebel & Welch (2022) <doi:10.1038/s41467-022-28431-4>.

Electrical properties of resistor networks using matrix methods.

Enhances the R Optimization Infrastructure ('ROI') package with the optimx package.

Includes algorithms to facilitate the assessment of extinction risk of species according to the IUCN (International Union for Conservation of Nature, see <https://iucn.org/> for more information) red list criteria.

We provide several avenues to predict and account for user-based mortality and tag loss during mark-recapture studies. When planning a study on a target species, the retentionmort_generation() function can be used to produce multiple synthetic mark-recapture datasets to anticipate the error associated with a planned field study to guide method development to reduce error. Similarly, if field data was already collected, the retentionmort() function can be used to predict the error from already generated data to adjust for user-based mortality and tag loss. The test_dataset_retentionmort() function will provide an example dataset of how data should be inputted into the function to run properly. Lastly, the retentionmort_figure() function can be used on any dataset generated from either model function to produce an rmarkdown printout of preliminary analysis associated with the model, including summary statistics and figures. Methods and results pertaining to the formation of this package can be found in McCutcheon et al. (in review, "Predicting tagging-related mortality and tag loss during mark-recapture studies").

Designed for longitudinal data analysis using Hidden Markov Models (HMMs). Tailored for applications in healthcare, social sciences, and economics, the main emphasis of this package is on regularization techniques for fitting HMMs. Additionally, it provides an implementation for fitting HMMs without regularization, referencing Zucchini et al. (2017, ISBN:9781315372488).

The implemented R6 class SCM aims to simplify working with structural causal models. The missing data mechanism can be defined as a part of the structural model. The class contains methods for 1) defining a structural causal model via functions, text or conditional probability tables, 2) printing basic information on the model, 3) plotting the graph for the model using packages igraph or qgraph', 4) simulating data from the model, 5) applying an intervention, 6) checking the identifiability of a query using the R packages causaleffect and dosearch', 7) defining the missing data mechanism, 8) simulating incomplete data from the model according to the specified missing data mechanism and 9) checking the identifiability in a missing data problem using the R package dosearch'. In addition, there are functions for running experiments and doing counterfactual inference using simulation.

Feasible multivariate GARCH models including DCC, GO-GARCH and Copula-GARCH.

This package provides a data structure and toolkit for documenting and recoding categorical data that can be shared in other statistical software.

Encrypt R objects to a raw vector or file using modern cryptographic techniques. Password-based key derivation is with Argon2 (<https://en.wikipedia.org/wiki/Argon2>). Objects are serialized and then encrypted using XChaCha20-Poly1305 (<https://en.wikipedia.org/wiki/ChaCha20-Poly1305>) which follows RFC 8439 for authenticated encryption (<https://en.wikipedia.org/wiki/Authenticated_encryption>). Cryptographic functions are provided by the included monocypher C library (<https://monocypher.org>).

Provide function for work with AcademyOcean API <https://academyocean.com/api>.

Extend Rasch and Item Response Theory (IRT) analyses by providing tools for post-processing the output from five major IRT packages (i.e., eRm', psychotools', ltm', mirt', and TAM'). The current version provides the plotPIccc() function, which extracts from the return object of the originating package all information required to draw an extended Person-Item-Map (PIccc), showing any combination of * category characteristic curves (CCCs), * threshold characteristic curves (TCCs), * item characteristic curves (ICCs), * category information functions (CIFs), * item information functions (IIFs), * test information function (TIF), and the * standard error curve (S.E.). for uni- and multidimensional models (as far as supported by each package). It allows for selecting dimensions, items, and categories to plot and offers numerous options to adapt the output. The return object contains all calculated values for further processing.

This package provides a collection of R functions for use with Stock Synthesis, a fisheries stock assessment modeling platform written in ADMB by Dr. Richard D. Methot at the NOAA Northwest Fisheries Science Center. The functions include tools for summarizing and plotting results, manipulating files, visualizing model parameterizations, and various other common stock assessment tasks. This version of r4ss is compatible with Stock Synthesis versions 3.24 through 3.30 (specifically version 3.30.19.01, from April 2022).

Implementations for several robust procedures that allow for (online) extraction of the signal of univariate or multivariate time series by applying robust regression techniques to a moving time window are provided. Included are univariate filtering procedures based on repeated-median regression as well as hybrid and trimmed filters derived from it; see Schettlinger et al. (2006) <doi:10.1515/BMT.2006.010>. The adaptive online repeated median by Schettlinger et al. (2010) <doi:10.1002/acs.1105> and the slope comparing adaptive repeated median by Borowski and Fried (2013) <doi:10.1007/s11222-013-9391-7> choose the width of the moving time window adaptively. Multivariate versions are also provided; see Borowski et al. (2009) <doi:10.1080/03610910802514972> for a multivariate online adaptive repeated median and Borowski (2012) <doi:10.17877/DE290R-14393> for a multivariate slope comparing adaptive repeated median. Furthermore, a repeated-median based filter with automatic outlier replacement and shift detection is provided; see Fried (2004) <doi:10.1080/10485250410001656444>.

Sundry discrete probability distributions and helper functions.

This package provides tools to evaluate the value of using a risk prediction instrument to decide treatment or intervention (versus no treatment or intervention). Given one or more risk prediction instruments (risk models) that estimate the probability of a binary outcome, rmda provides functions to estimate and display decision curves and other figures that help assess the population impact of using a risk model for clinical decision making. Here, "population" refers to the relevant patient population. Decision curves display estimates of the (standardized) net benefit over a range of probability thresholds used to categorize observations as high risk'. The curves help evaluate a treatment policy that recommends treatment for patients who are estimated to be high risk by comparing the population impact of a risk-based policy to "treat all" and "treat none" intervention policies. Curves can be estimated using data from a prospective cohort. In addition, rmda can estimate decision curves using data from a case-control study if an estimate of the population outcome prevalence is available. Version 1.4 of the package provides an alternative framing of the decision problem for situations where treatment is the standard-of-care and a risk model might be used to recommend that low-risk patients (i.e., patients below some risk threshold) opt out of treatment. Confidence intervals calculated using the bootstrap can be computed and displayed. A wrapper function to calculate cross-validated curves using k-fold cross-validation is also provided.

The Agricultural Production Systems sIMulator ('APSIM') is a widely used to simulate the agricultural systems for multiple crops. This package is designed to create, modify and run apsimx files in the APSIM Next Generation <https://www.apsim.info/>.

Fast computing an ensemble of rank-based trees via boosting or random forest on binary and multi-class problems. It converts continuous gene expression profiles into ranked gene pairs, for which the variable importance indices are computed and adopted for dimension reduction. Decision rules can be extracted from trees.

Bootstrap forecast densities for GARCH (Generalized Autoregressive Conditional Heteroskedastic) returns and volatilities using the robust residual-based bootstrap procedure of Trucios, Hotta and Ruiz (2017) <DOI:10.1080/00949655.2017.1359601>.

The method generate() is extended for spatial multi-site stochastic generation of daily precipitation. It generates precipitation occurrence in several sites using logit regression (Generalized Linear Models) and the approach by D.S. Wilks (1998) <doi:10.1016/S0022-1694(98)00186-3> .

Allows the user to generate and execute select, insert, update and delete SQL queries the underlying database without having to explicitly write SQL code.

Tu & Zhou (1999) <doi:10.1002/(SICI)1097-0258(19991030)18:20%3C2749::AID-SIM195%3E3.0.CO;2-C> showed that comparing the means of populations whose data-generating distributions are non-negative with excess zero observations is a problem of great importance in the analysis of medical cost data. In the same study, Tu & Zhou discuss that it can be difficult to control type-I error rates of general-purpose statistical tests for comparing the means of these particular data sets. This package allows users to perform a modified bootstrap-based t-test that aims to better control type-I error rates in these situations.

Use trend filtering, a type of regularized nonparametric regression, to estimate the instantaneous reproduction number, also called Rt. This value roughly says how many new infections will result from each new infection today. Values larger than 1 indicate that an epidemic is growing while those less than 1 indicate decline. For more details about this methodology, see Liu, Cai, Gustafson, and McDonald (2024) <doi:10.1371/journal.pcbi.1012324>.