Instrumental variables (IVs) are a popular and powerful tool for estimating causal effects in the presence of unobserved confounding. However, classical methods rely on strong assumptions such as the exclusion criterion, which states that instrumental effects must be entirely mediated by treatments. In the so-called "leaky" IV setting, candidate instruments are allowed to have some direct influence on outcomes, rendering the average treatment effect (ATE) unidentifiable. But with limits on the amount of information leakage, we may still recover sharp bounds on the ATE, providing partial identification. This package implements methods for ATE bounding in the leaky IV setting with linear structural equations. For details, see Watson et al. (2024) <doi:10.48550/arXiv.2404.04446>.

Monte Carlo confidence intervals for free and defined parameters in models fitted in the structural equation modeling package lavaan can be generated using the semmcci package. semmcci has three main functions, namely, MC(), MCMI(), and MCStd(). The output of lavaan is passed as the first argument to the MC() function or the MCMI() function to generate Monte Carlo confidence intervals. Monte Carlo confidence intervals for the standardized estimates can also be generated by passing the output of the MC() function or the MCMI() function to the MCStd() function. A description of the package and code examples are presented in Pesigan and Cheung (2024) <doi:10.3758/s13428-023-02114-4>.

This package provides visualizations for SHAP (SHapley Additive exPlanations) such as waterfall plots, force plots, various types of importance plots, dependence plots, and interaction plots. These plots act on a shapviz object created from a matrix of SHAP values and a corresponding feature dataset. Wrappers for the R packages xgboost, lightgbm, fastshap, shapr, h2o, treeshap, DALEX, and kernelshap are added for convenience. By separating visualization and computation, it is possible to display factor variables in graphs, even if the SHAP values are calculated by a model that requires numerical features. The plots are inspired by those provided by the shap package in Python, but there is no dependency on it.

This package provides users with its associated functions for pedagogical purposes in visually learning Bayesian networks and Markov chain Monte Carlo (MCMC) computations. It enables users to: a) Create and examine the (starting) graphical structure of Bayesian networks; b) Create random Bayesian networks using a dataset with customized constraints; c) Generate Stan code for structures of Bayesian networks for sampling the data and learning parameters; d) Plot the network graphs; e) Perform Markov chain Monte Carlo computations and produce graphs for posteriors checks. The package refers to one reference item, which describes the methods and algorithms: Vuong, Quan-Hoang and La, Viet-Phuong (2019) <doi:10.31219/osf.io/w5dx6> The bayesvl R package. Open Science Framework (May 18).

This package provides R utilities to build unlevered and levered discounted cash flow (DCF) tables for commercial real estate (CRE) assets. Functions generate bullet and amortising debt schedules, compute credit metrics such as debt service coverage ratios (DSCR), debt yield ratios, and forward loan-to-value ratios (LTV), and expose an explicit property-level operating chain from gross effective income (GEI) to net operating income (NOI) and property before-tax cash flow (PBTCF). The toolkit supports end-to-end scenario execution from a YAML (YAML Ain't Markup Language) configuration file parsed with yaml', includes helpers for effective rent, constrained loan underwriting, and simplified SPV-level tax simulations, and ships reproducible vignettes for methodological and applied use cases.

Compute price indices using various Hedonic and multilateral methods, including Laspeyres, Paasche, Fisher, and HMTS (Hedonic Multilateral Time series re-estimation with splicing). The central function calculate_price_index() offers a unified interface for running these methods on structured datasets. This package is designed to support index construction workflows for real estate and other domains where quality-adjusted price comparisons over time are essential. The development of this package was funded by Eurostat and Statistics Netherlands (CBS), and carried out by Statistics Netherlands. The HMTS method implemented here is described in Ishaak, Ouwehand and Remøy (2024) <doi:10.1177/0282423X241246617>. For broader methodological context, see Eurostat (2013, ISBN:978-92-79-25984-5, <doi:10.2785/34007>).

Fits or generalized linear models either a regression with Autoregressive moving-average (ARMA) errors for time series data. The package makes it easy to incorporate constraints into the model's coefficients. The model is specified by an objective function (Gaussian, Binomial or Poisson) or an ARMA order (p,q), a vector of bound constraints for the coefficients (i.e beta1 > 0) and the possibility to incorporate restrictions among coefficients (i.e beta1 > beta2). The references of this packages are the same as stats package for glm() and arima() functions. See Brockwell, P. J. and Davis, R. A. (1996, ISBN-10: 9783319298528). For the different optimizers implemented, it is recommended to consult the documentation of the corresponding packages.

An interface to DifferentialEquations.jl <https://diffeq.sciml.ai/dev/> from the R programming language. It has unique high performance methods for solving ordinary differential equations (ODE), stochastic differential equations (SDE), delay differential equations (DDE), differential-algebraic equations (DAE), and more. Much of the functionality, including features like adaptive time stepping in SDEs, are unique and allow for multiple orders of magnitude speedup over more common methods. Supports GPUs, with support for CUDA (NVIDIA), AMD GPUs, Intel oneAPI GPUs, and Apple's Metal (M-series chip GPUs). diffeqr attaches an R interface onto the package, allowing seamless use of this tooling by R users. For more information, see Rackauckas and Nie (2017) <doi:10.5334/jors.151>.

This package provides a unified interface to fit, cross-validate and benchmark genomic prediction models from SNP marker data. It implements genomic best linear unbiased prediction (GBLUP) and ridge-regression BLUP in base R, and offers a common interface to machine-learning predictors (elastic net, random forest and gradient boosting) through optional packages, together with a stacked ensemble. Cross-validation uses breeding-relevant schemes and reports prediction accuracy honestly, so models can be compared fairly. The genomic relationship matrix follows VanRaden (2008) <doi:10.3168/jds.2007-0980>; the mixed-model solver follows Endelman (2011) <doi:10.3835/plantgenome2011.08.0024>; the genomic-selection framework follows Meuwissen, Hayes and Goddard (2001) <doi:10.1093/genetics/157.4.1819>.

Converts TXT and XML data curated by the United States Patent and Trademark Office (USPTO). Allows conversion of bulk data after downloading directly from the USPTO bulk data website, eliminating need for users to wrangle multiple data formats to get large patent databases in tidy, rectangular format. Data details can be found on the USPTO website <https://bulkdata.uspto.gov/>. Currently, all 3 formats: 1. TXT data (1976-2001); 2. XML format 1 data (2002-2004); and 3. XML format 2 data (2005-current) can be converted to rectangular, CSV format. Relevant literature that uses data from USPTO includes Wada (2020) <doi:10.1007/s11192-020-03674-4> and Plaza & Albert (2008) <doi:10.1007/s11192-007-1763-3>.

This package implements the Savvy Parity Regression savvyPR methodology for multivariate linear regression analysis. The package solves an optimization problem that balances the contribution of each predictor variable to ensure estimation stability in the presence of multicollinearity. It supports two distinct parameterization methods, a Budget-based approach that allocates a fixed loss contribution to each predictor, and a Target-based approach (t-tuning) that utilizes a relative elasticity weight for the response variable. The package provides comprehensive tools for model estimation, risk distribution analysis, and parameter tuning via cross-validation (PR1, PR2, and PR3 model types) to optimize predictive accuracy. Methods are based on Asimit, Chen, Ichim and Millossovich (2026) <https://openaccess.city.ac.uk/id/eprint/37017/>.

This package provides raw files recorded on different Liquid Chromatography Mass Spectrometry (LC-MS) instruments. All included MS instruments are manufactured by Thermo Fisher Scientific and belong to the Orbitrap Tribrid or Q Exactive Orbitrap family of instruments. Despite their common origin and shared hardware components, e.g., Orbitrap mass analyser, the above instruments tend to write data in different "dialects" in a shared binary file format (.raw). The intention behind tartare is to provide complex but slim real-world files that can be used to make code robust with respect to this diversity. In other words, it is intended for enhanced unit testing. The package is considered to be used with the rawrr package and the Spectra MsBackends.

For spatial data analysis; provides exploratory spatial analysis tools, spatial regression, spatial econometric, and disease mapping models, model diagnostics, and special methods for inference with small area survey data (e.g., the America Community Survey (ACS)) and censored population health monitoring data. Models are pre-specified using the Stan programming language, a platform for Bayesian inference using Markov chain Monte Carlo (MCMC). References: Carpenter et al. (2017) <doi:10.18637/jss.v076.i01>; Donegan (2021) <doi:10.31219/osf.io/3ey65>; Donegan (2022) <doi:10.21105/joss.04716>; Donegan, Chun and Hughes (2020) <doi:10.1016/j.spasta.2020.100450>; Donegan, Chun and Griffith (2021) <doi:10.3390/ijerph18136856>; Morris et al. (2019) <doi:10.1016/j.sste.2019.100301>.

This package implements Latent Unknown Clusters By Integrating Multi-omics Data (LUCID; Peng (2019) <doi:10.1093/bioinformatics/btz667>) for integrative clustering with exposures, multi-omics data, and health outcomes. Supports three integration strategies: early, parallel, and serial. Provides model fitting and tuning, lasso-type regularization for exposure and omics feature selection, handling of missing data, including both sporadic and complete-case patterns, prediction, and g-computation for estimating causal effects of exposures, bootstrap inference for uncertainty estimation, and S3 summary and plot methods. For the multi-omics integration framework, see Jia (2024) <https://journal.r-project.org/articles/RJ-2024-012/RJ-2024-012.pdf>. For the missing-data imputation mechanism, see Jia (2024) <doi:10.1093/bioadv/vbae123>.

SEM Trees and SEM Forests -- an extension of model-based decision trees and forests to Structural Equation Models (SEM). SEM trees hierarchically split empirical data into homogeneous groups each sharing similar data patterns with respect to a SEM by recursively selecting optimal predictors of these differences. SEM forests are an extension of SEM trees. They are ensembles of SEM trees each built on a random sample of the original data. By aggregating over a forest, we obtain measures of variable importance that are more robust than measures from single trees. A description of the method was published by Brandmaier, von Oertzen, McArdle, & Lindenberger (2013) <doi:10.1037/a0030001> and Arnold, Voelkle, & Brandmaier (2020) <doi:10.3389/fpsyg.2020.564403>.

This package provides a collection of functions for estimating spatial and spatio-temporal regression models. Moran eigenvectors are used as spatial basis functions to efficiently approximate spatially dependent Gaussian processes (i.e., random effects eigenvector spatial filtering; see Murakami and Griffith 2015 <doi: 10.1007/s10109-015-0213-7>). The implemented models include linear regression with residual spatial dependence, spatially/spatio-temporally varying coefficient models (Murakami et al., 2017, 2024; <doi:10.1016/j.spasta.2016.12.001>,<doi:10.48550/arXiv.2410.07229>), spatially filtered unconditional quantile regression (Murakami and Seya, 2019 <doi:10.1002/env.2556>), Gaussian and non-Gaussian spatial mixed models through compositionally-warping (Murakami et al. 2021, <doi:10.1016/j.spasta.2021.100520>).

This package provides functions for building cognitive maps based on qualitative data. Inputs are textual sources (articles, transcription of qualitative interviews of agents,...). These sources have been coded using relations and are linked to (i) a table describing the variables (or concepts) used for the coding and (ii) a table describing the sources (typology of agents, ...). Main outputs are Individual Cognitive Maps (ICM), Social Cognitive Maps (all sources or group of sources) and a list of quotes linked to relations. This package is linked to the work done during the PhD of Frederic M. Vanwindekens (CRA-W / UCL) hold the 13 of May 2014 at University of Louvain in collaboration with the Walloon Agricultural Research Centre (project MIMOSA, MOERMAN fund).

"Evolutionary Virtual Education" - evolved - provides multiple tools to help educators (especially at the graduate level or in advanced undergraduate level courses) apply inquiry-based learning in general evolution classes. In particular, the tools provided include functions that simulate evolutionary processes (e.g., genetic drift, natural selection within a single locus) or concepts (e.g. Hardy-Weinberg equilibrium, phylogenetic distribution of traits). More than only simulating, the package also provides tools for students to analyze (e.g., measuring, testing, visualizing) datasets with characteristics that are common to many fields related to evolutionary biology. Importantly, the package is heavily oriented towards providing tools for inquiry-based learning - where students follow scientific practices to actively construct knowledge. For additional details, see package's vignettes.

Deconvolution of thermal decay curves allows you to quantify proportions of biomass components in plant litter. Thermal decay curves derived from thermogravimetric analysis (TGA) are imported, modified, and then modelled in a three- or four- part mixture model using the Fraser-Suzuki function. The output is estimates for weights of pseudo-components corresponding to hemicellulose, cellulose, and lignin. For more information see: Müller-Hagedorn, M. and Bockhorn, H. (2007) <doi:10.1016/j.jaap.2006.12.008>, à rfão, J. J. M. and Figueiredo, J. L. (2001) <doi:10.1016/S0040-6031(01)00634-7>, and Yang, H. and Yan, R. and Chen, H. and Zheng, C. and Lee, D. H. and Liang, D. T. (2006) <doi:10.1021/ef0580117>.

This package provides tools for calculating and visualizing maritime distances and routes between geographic points. At its core, it implements a fast Haversine formula implemented in data.table to compute great circle distances across sea regions (i.e. avoiding land mass). The package builds a spatial network graph from port and cluster coordinates and uses a shortest path algorithm to identify optimal maritime routes between origin-destination pairs. For visualization, the package exports maps displaying individual routes, multi-destination networks, or continuous routes through specified waypoints. Utility functions identify the nearest network nodes to arbitrary coordinates and handle the antimeridian discontinuities common in Pacific maritime mapping. The package is particularly suited for analyzing shipping lanes, trade routes, and vessel trajectory data.

Computes sample size and power for causal inference studies that use propensity score (PS) weighting. Supports continuous, binary, and time-to-event (survival) outcomes under four estimands: average treatment effect (ATE), average treatment effect on the treated (ATT), average treatment effect on the controls (ATC), and average treatment effect on the overlap population (ATO). For continuous and binary outcomes, the asymptotic variance of the Hajek inverse probability weighting estimator is derived under a logit-normal propensity score model, approximated by a Beta distribution matched through the Bhattacharyya overlap coefficient. For survival outcomes, the asymptotic variance of the propensity-score- weighted partial likelihood estimator is used for randomized trials and observational studies. The Schoenfeld formula is also available for randomized trial settings.

Genotyping arrays enable the direct measurement of an individuals genotype at thousands of markers. plinkQC facilitates genotype quality control for genetic association studies as described by Anderson and colleagues (2010) <doi:10.1038/nprot.2010.116>. It makes PLINK basic statistics (e.g. missing genotyping rates per individual, allele frequencies per genetic marker) and relationship functions accessible from R and generates a per-individual and per-marker quality control report. Individuals and markers that fail the quality control can subsequently be removed to generate a new, clean dataset. Removal of individuals based on relationship status is optimised to retain as many individuals as possible in the study. Additionally, there is a trained classifier to predict genomic ancestry of human samples.

This package provides semiparametric smooth-coefficient stochastic frontier analysis following Sun and Kumbhakar (2013) <doi:10.1016/j.econlet.2013.05.001> where the coefficients of the parametric part vary smoothly with a set of nonparametric variables. Inefficiency term is allowed to depend on a set of determinants through heteroskedasticity. Smooth coefficients are estimated using nonparametric regression and the remaining frontier parameters are estimated by maximum likelihood. Technical efficiency and inefficiency are computed using the Battese and Coelli (1988) <doi:10.1016/0304-4076(88)90053-X> and Jondrow et al. (1982) <doi:10.1016/0304-4076(82)90004-5> methods, respectively. Confidence intervals for technical efficiency are computed using the approach of Horrace and Schmidt (1996) <doi:10.1007/BF00157044>.

This package provides a framework of tools to summarise, visualise, and explore longitudinal data. It builds upon the tidy time series data frames used in the tsibble package, and is designed to integrate within the tidyverse', and tidyverts (for time series) ecosystems. The methods implemented include calculating features for understanding longitudinal data, including calculating summary statistics such as quantiles, medians, and numeric ranges, sampling individual series, identifying individual series representative of a group, and extending the facet system in ggplot2 to facilitate exploration of samples of data. These methods are fully described in the paper "brolgar: An R package to Browse Over Longitudinal Data Graphically and Analytically in R", Nicholas Tierney, Dianne Cook, Tania Prvan (2020) <doi:10.32614/RJ-2022-023>.