Approximate node interaction parameters of Markov Random Fields graphical networks. Models can incorporate additional covariates, allowing users to estimate how interactions between nodes in the graph are predicted to change across covariate gradients. The general methods implemented in this package are described in Clark et al. (2018) <doi:10.1002/ecy.2221>.

Describes spatial patterns of categorical raster data for any defined regular and irregular areas. Patterns are described quantitatively using built-in signatures based on co-occurrence matrices but also allows for any user-defined functions. It enables spatial analysis such as search, change detection, and clustering to be performed on spatial patterns (Nowosad (2021) <doi:10.1007/s10980-020-01135-0>).

Implementation of Matched Wake Analysis (mwa) for studying causal relationships in spatiotemporal event data, introduced by Schutte and Donnay (2014) <doi:10.1016/j.polgeo.2014.03.001>.

Load, validate, and manipulate Clinical Data Interchange Standards Consortium ('CDISC') Analysis Data Model ('ADaM') dataset metadata stored as YAML files. Metadata files are validated against a JSON schema. Provides functions to inspect and modify columns, parameters, and row-level operations within and across ADaM domains. Designed for use with the mighty framework.

Estimates membership for the Mandelbrot set.

Chromatin immunoprecipitation followed by high-throughput sequencing (ChIP-seq) is the premier technology for profiling genome-wide localization of chromatin-binding proteins, including transcription factors and histones with various modifications. This package provides a robust method for normalizing ChIP-seq signals across individual samples or groups of samples. It also designs a self-contained system of statistical models for calling differential ChIP-seq signals between two or more biological conditions as well as for calling hypervariable ChIP-seq signals across samples. Refer to Tu et al. (2021) <doi:10.1101/gr.262675.120> and Chen et al. (2022) <doi:10.1186/s13059-022-02627-9> for associated statistical details.

Quantify exposure-outcome causal effects with adjustment for multiple biases. The functions can simultaneously adjust for any combination of uncontrolled confounding, exposure/outcome misclassification, and selection bias. The underlying method generalizes the combination of inverse probability of selection weighting with predictive value weighting. Simultaneous multi-bias analysis can be used to enhance the validity and transparency of real-world evidence obtained from observational, longitudinal studies. Based on the work from Paul Brendel, Aracelis Torres, and Onyebuchi Arah (2023) <doi:10.1093/ije/dyad001>.

An implementation of the mixed neighbourhood selection (MNS) algorithm. The MNS algorithm can be used to estimate multiple related precision matrices. In particular, the motivation behind this work was driven by the need to understand functional connectivity networks across multiple subjects. This package also contains an implementation of a novel algorithm through which to simulate multiple related precision matrices which exhibit properties frequently reported in neuroimaging analysis.

Auto-downloads Parquet data from the MTGJSON CDN and exposes the full Magic: The Gathering dataset through R6-based query interfaces backed by DuckDB'.

This package implements several methods to meta-analyze studies that report the sample median of the outcome. The methods described by McGrath et al. (2019) <doi:10.1002/sim.8013>, Ozturk and Balakrishnan (2020) <doi:10.1002/sim.8738>, and McGrath et al. (2020a) <doi:10.1002/bimj.201900036> can be applied to directly meta-analyze the median or difference of medians between groups. Additionally, a number of methods (e.g., McGrath et al. (2020b) <doi:10.1177/0962280219889080>, Cai et al. (2021) <doi:10.1177/09622802211047348>, and McGrath et al. (2023) <doi:10.1177/09622802221139233>) are implemented to estimate study-specific (difference of) means and their standard errors in order to estimate the pooled (difference of) means. Methods for meta-analyzing median survival times (McGrath et al. (2026) <doi:10.1002/sim.70533>) are also implemented. See McGrath et al. (2024) <doi:10.1002/jrsm.1686> for a detailed guide on using the package.

This package provides a generalization of the Synth package that is designed for data at a more granular level (e.g., micro-level). Provides functions to construct weights (including propensity score-type weights) and run analyses for synthetic control methods with micro- and meso-level data; see Robbins, Saunders, and Kilmer (2017) <doi:10.1080/01621459.2016.1213634> and Robbins and Davenport (2021) <doi:10.18637/jss.v097.i02>.

Maximum a posteriori (MAP) estimation for topic models (i.e., Latent Dirichlet Allocation) in text analysis, as described in Taddy (2012) On estimation and selection for topic models'. Previous versions of this code were included as part of the textir package. If you want to take advantage of openmp parallelization, uncomment the relevant flags in src/MAKEVARS before compiling.

Uses multiple AUCs to select a combination of predictors when the outcome has multiple (ordered) levels and the focus is discriminating one particular level from the others. This method is most naturally applied to settings where the outcome has three levels. (Meisner, A, Parikh, CR, and Kerr, KF (2017) <http://biostats.bepress.com/uwbiostat/paper423/>.).

Computes efficient data distributions from highly inconsistent datasets with many missing values using multi-set intersections. Based upon hash functions, mulset can quickly identify intersections from very large matrices of input vectors across columns and rows and thus provides scalable solution for dealing with missing values. Tomic et al. (2019) <doi:10.1101/545186>.

This package provides a mechanism to plot an interactive map using Mapbox GL (<https://docs.mapbox.com/mapbox-gl-js/api/>), a javascript library for interactive maps, and Deck.gl (<https://deck.gl/>), a javascript library which uses WebGL for visualising large data sets.

The MIDASim package is a microbiome data simulator for generating realistic microbiome datasets by adapting a user-provided template. It supports the controlled introduction of experimental signals-such as shifts in taxon relative abundances, prevalence, and sample library sizes-to create distinct synthetic populations under diverse simulation scenarios. For more details, see He et al. (2024) <doi:10.1186/s40168-024-01822-z>.

Package for estimating, analyzing, and forecasting multi-country macro-finance affine term structure models (ATSMs). All setups build on the single-country unspanned macroeconomic risk framework from Joslin, Priebsch, and Singleton (2014, JF) <doi:10.1111/jofi.12131>. Multicountry extensions by Jotikasthira, Le, and Lundblad (2015, JFE) <doi:10.1016/j.jfineco.2014.09.004>, Candelon and Moura (2023, EM) <doi:10.1016/j.econmod.2023.106453>, and Candelon and Moura (2024, JFEC) <doi:10.1093/jjfinec/nbae008> are also available. The package also provides tools for bias correction as in Bauer Rudebusch and Wu (2012, JBES) <doi:10.1080/07350015.2012.693855>, bootstrap analysis, and several graphical/numerical outputs.

Forecasting of univariate time series using feature extraction with variable prediction methods is provided. Feature extraction is done with a redundant Haar wavelet transform with filter h = (0.5, 0.5). The advantage of the approach compared to typical Fourier based methods is an dynamic adaptation to varying seasonalities. Currently implemented prediction methods based on the selected wavelets levels and scales are a regression and a multi-layer perceptron. Forecasts can be computed for horizon 1 or higher. Model selection is performed with an evolutionary optimization. Selection criteria are currently the AIC criterion, the Mean Absolute Error or the Mean Root Error. The data is split into three parts for model selection: Training, test, and evaluation dataset. The training data is for computing the weights of a parameter set. The test data is for choosing the best parameter set. The evaluation data is for assessing the forecast performance of the best parameter set on new data unknown to the model. This work is published in Stier, Q.; Gehlert, T.; Thrun, M.C. Multiresolution Forecasting for Industrial Applications. Processes 2021, 9, 1697. <doi:10.3390/pr9101697>.

This package provides methods for analyzing and using quartets displayed on a collection of gene trees, primarily to make inferences about the species tree or network under the multi-species coalescent model. These include quartet hypothesis tests for the model, as developed by Mitchell et al. (2019) <doi:10.1214/19-EJS1576>, simplex plots of quartet concordance factors as presented by Allman et al. (2020) <doi:10.1101/2020.02.13.948083>, species tree inference methods based on quartet distances of Rhodes (2019) <doi:10.1109/TCBB.2019.2917204> and Yourdkhani and Rhodes (2019) <doi:10.1007/s11538-020-00773-4>, the NANUQ algorithm for inference of level-1 species networks of Allman et al. (2019) <doi:10.1186/s13015-019-0159-2>, the TINNIK algorithm for inference of the tree of blobs of an arbitrary network of Allman et al.(2022) <doi:10.1007/s00285-022-01838-9>, NANUQ+ routines for resolving multifurcations in the tree of blobs to cycles as in Rhodes et al.(2024) <doi:10.1186/s13015-025-00274-w>, and the ECToBlob algorithm for inference of a network with no anomalous quartets of Allman et al. (2026) (forthcoming). Software announcement by Rhodes et al. (2020) <doi:10.1093/bioinformatics/btaa868>.

The companion package provides all original data sets and functions that are used in the book "Model-Based Clustering and Classification for Data Science" by Charles Bouveyron, Gilles Celeux, T. Brendan Murphy and Adrian E. Raftery (2019, ISBN:9781108644181).

Measure of the Effect ('MOTE') is an effect size calculator, including a wide variety of effect sizes in the mean differences family (all versions of d) and the variance overlap family (eta, omega, epsilon, r). MOTE provides non-central confidence intervals for each effect size, relevant test statistics, and output for reporting in APA Style (American Psychological Association, 2010, <ISBN:1433805618>) with LaTeX'. In research, an over-reliance on p-values may conceal the fact that a study is under-powered (Halsey, Curran-Everett, Vowler, & Drummond, 2015 <doi:10.1038/nmeth.3288>). A test may be statistically significant, yet practically inconsequential (Fritz, Scherndl, & Kühberger, 2012 <doi:10.1177/0959354312436870>). Although the American Psychological Association has long advocated for the inclusion of effect sizes (Wilkinson & American Psychological Association Task Force on Statistical Inference, 1999 <doi:10.1037/0003-066X.54.8.594>), the vast majority of peer-reviewed, published academic studies stop short of reporting effect sizes and confidence intervals (Cumming, 2013, <doi:10.1177/0956797613504966>). MOTE simplifies the use and interpretation of effect sizes and confidence intervals.

Conducts one- and two-sample hypothesis tests for median absolute deviations (mads) for robust inference of dispersion. Comparisons between two samples uses the ratio of mads. Confidence intervals are also computed.

Fits generalized linear models, Cox proportional-hazards models, log-rank tests, generalized estimating equations, mixed models with repeated measures, Kaplan-Meier curves, quantile differences, and hierarchical net-benefit (win-difference) and log win-ratio statistics jointly across multiple endpoints, and returns the full asymptotic covariance matrix linking them. Implements PATED (Prognostic Assisted Treatment Effect Detection), a randomized-trial method that exploits balanced prognostic covariates to tighten standard errors and increase statistical power without introducing bias.

Implementation of Warnes & Raftery's MCGibbsit run-length and convergence diagnostic for a set of (not-necessarily independent) Markov Chain Monte Carlo (MCMC) samplers. It combines the quantile estimate error-bounding approach of the Raftery and Lewis MCMC run length diagnostic `gibbsit` with the between verses within chain approach of the Gelman and Rubin MCMC convergence diagnostic.