Loads, validates, and replays portable mars ModelSpec artifacts from R. The package provides helpers for constructing design matrices, generating predictions, and, when the companion runtime helper is available, fitting portable model specifications for cross-language replay.

This package provides a collection of functions to do some statistical inferences. On estimation, it has the function to get the method of moments estimates, the sampling interval. In terms of testing it has function of doing most powerful test.

This package provides graph-constrained regression methods in which regularization parameters are selected automatically via estimation of equivalent Linear Mixed Model formulation. riPEER (ridgified Partially Empirical Eigenvectors for Regression) method employs a penalty term being a linear combination of graph-originated and ridge-originated penalty terms, whose two regularization parameters are ML estimators from corresponding Linear Mixed Model solution; a graph-originated penalty term allows imposing similarity between coefficients based on graph information given whereas additional ridge-originated penalty term facilitates parameters estimation: it reduces computational issues arising from singularity in a graph-originated penalty matrix and yields plausible results in situations when graph information is not informative. riPEERc (ridgified Partially Empirical Eigenvectors for Regression with constant) method utilizes addition of a diagonal matrix multiplied by a predefined (small) scalar to handle the non-invertibility of a graph Laplacian matrix. vrPEER (variable reducted PEER) method performs variable-reduction procedure to handle the non-invertibility of a graph Laplacian matrix.

We introduce a high-dimensional multi-study robust factor model, which learns latent features and accounts for the heterogeneity among source. It could be used for analyzing heterogeneous RNA sequencing data. More details can be referred to Jiang et al. (2025) <doi:10.48550/arXiv.2506.18478>.

Takes QC signal for each day and normalize metabolomic data that has been acquired in a certain period of time. At least three QC per day are required.

Computes model-based effect sizes for fixed-effect coefficients in multilevel (hierarchical) models. The coefficient effect sizes are standardized mean differences from zero (d) and unique variance-explained measures (squared semi-partial correlations, sr2). The package also reports variance components and level-specific and total R-squared values. It supports 2-level and 3-level linear and binary logistic models fitted with lme4 (Bates et al., 2015) <doi:10.18637/jss.v067.i01>, and 2-level Gaussian and Bernoulli models fitted with brms (Bürkner, 2017) <doi:10.18637/jss.v080.i01>. Sanders, Konold, and Cheng (in press), "Model-based effect sizes for multilevel linear regression coefficients," Methodology: European Journal of Research Methods for the Behavioral and Social Sciences, describe the 2-level linear-model methods.

Generate multivariate discrete data with generalized Poisson, negative binomial and binomial marginal distributions using user-specified distribution parameters and a target correlation matrix. The method is described in Cheng and Demirtas (2026) <doi:10.48550/arXiv.2602.07707>.

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/>.).

Incorporates Approximate Bayesian Computation to get a posterior distribution and to select a model optimal parameter for an observation point. Additionally, the meta-sampling heuristic algorithm is realized for parameter estimation, which requires no model runs and is dimension-independent. A sampling scheme is also presented that allows model runs and uses the meta-sampling for point generation. A predictor is realized as the meta-sampling for the model output. All the algorithms leverage a machine learning method utilizing the maxima weighted Isolation Kernel approach, or MaxWiK'. The method involves transforming raw data to a Hilbert space (mapping) and measuring the similarity between simulated points and the maxima weighted Isolation Kernel mapping corresponding to the observation point. Comprehensive details of the methodology can be found in the papers Iurii Nagornov (2024) <doi:10.1007/978-3-031-66431-1_16> and Iurii Nagornov (2023) <doi:10.1007/978-3-031-29168-5_18>.

Shiny web application to run meta-analyses. Essentially a graphical front-end to package meta for R. Can be useful as an educational tool, and for quickly analyzing and sharing meta-analyses. Provides output to quickly fill in GRADE (Grading of Recommendations, Assessment, Development and Evaluations) Summary-of-Findings tables. Importantly, it allows further processing of the results inside R, in case more specific analyses are needed.

Multi-step adaptive elastic-net (MSAENet) algorithm for feature selection in high-dimensional regressions proposed in Xiao and Xu (2015) <DOI:10.1080/00949655.2015.1016944>, with support for multi-step adaptive MCP-net (MSAMNet) and multi-step adaptive SCAD-net (MSASNet) methods.

Early insights in probability theory were largely influenced by questions about gambling and games of chance, as noted by Blitzstein and Hwang (2019, ISBN:978-1138369917). In modern times, playing cards continue to serve as an effective teaching tool for probability, statistics, and even R programming, as demonstrated by Grolemund (2014, ISBN:978-1449359010). The mmcards package offers a collection of utility functions designed to aid in the creation, manipulation, and utilization of playing card decks in multiple formats. These include a standard 52-card deck, as well as alternative decks such as decks defined by custom anonymous functions and custom interleaved decks. Optimized for the development of educational shiny applications, the package is particularly useful for teaching statistics and probability through card-based games. Functions include shuffle_deck(), which creates either a shuffled standard deck or a shuffled custom alternative deck; deal_card(), which takes a deck and returns a list object containing both the dealt card and the updated deck; and i_deck(), which adds image paths to card objects, further enriching the package's utility in the development of interactive shiny application card games.

Translate R code into MongoDB aggregation pipelines.

The main function performs meta-analysis of genetic association study summary statistics that may be correlated due to cryptic relatedness or other confounders, generalizing inverse variance weighted methods. The function that estimates the correlation structure is also provided standalone. Another key innovation, the estimation of the correlation parameter from the median product of correlated standard normal variables, is provided, as well as a complete set of functions for their underlying distribution: density, cumulative, quantile, and random deviates. Described in Tu and Ochoa (2026) <doi:10.1016/j.xhgg.2026.100657>.

Constructs a space-filling design under the criterion of maximum-minimum distance. Both discrete and continuous searches are provided.

This repository aims to contribute to the econometric models production with Colombian data, by providing a set of web-scrapping functions of some of the main macro-financial indicators. All the sources are public and free, but the advantage of these functions is that they directly download and harmonize the information in R's environment. No need to import or download additional files. You only need an internet connection!

Simulate, manage, visualize, and analyze spatially and temporally explicit datasets of mating potential. Implements methods to calculate synchrony, proximity, and compatibility.Synchrony calculations are based on methods described in Augspurger (1983) <doi:10.2307/2387650>, Kempenaers (1993) <doi:10.2307/3676415>, Ison et al. (2014) <doi:10.3732/ajb.1300065>, and variations on these, as described.

This package performs meta-analysis and meta-regression using standard and robust methods with confidence intervals based on the profile likelihood. Robust methods are based on alternative distributions for the random effect, either the t-distribution (Lee and Thompson, 2008 <doi:10.1002/sim.2897> or Baker and Jackson, 2008 <doi:10.1007/s10729-007-9041-8>) or mixtures of normals (Beath, 2014 <doi:10.1002/jrsm.1114>).

Matrix eQTL is designed for fast eQTL analysis on large datasets. Matrix eQTL can test for association between genotype and gene expression using linear regression with either additive or ANOVA genotype effects. The models can include covariates to account for factors as population stratification, gender, and clinical variables. It also supports models with heteroscedastic and/or correlated errors, false discovery rate estimation and separate treatment of local (cis) and distant (trans) eQTLs. For more details see Shabalin (2012) <doi:10.1093/bioinformatics/bts163>.

Bayesian estimation of inverse variance weighted (IVW), Burgess et al. (2013) <doi:10.1002/gepi.21758>, and MR-Egger, Bowden et al. (2015) <doi:10.1093/ije/dyv080>, summary data models for Mendelian randomization analyses.

Straightforward and detailed evaluation of machine learning models. MLeval can produce receiver operating characteristic (ROC) curves, precision-recall (PR) curves, calibration curves, and PR gain curves. MLeval accepts a data frame of class probabilities and ground truth labels, or, it can automatically interpret the Caret train function results from repeated cross validation, then select the best model and analyse the results. MLeval produces a range of evaluation metrics with confidence intervals.

An increasingly important source of health-related bibliographic content are preprints - preliminary versions of research articles that have yet to undergo peer review. The two preprint repositories most relevant to health-related sciences are medRxiv <https://www.medrxiv.org/> and bioRxiv, both of which are operated by the Cold Spring Harbor Laboratory. medrxivr provides programmatic access to the Cold Spring Harbour Laboratory (CSHL) API <https://api.biorxiv.org/>, allowing users to easily download medRxiv and bioRxiv preprint metadata (e.g. title, abstract, publication date, author list, etc) into R. medrxivr also provides functions to search the downloaded preprint records using regular expressions and Boolean logic, as well as helper functions that allow users to export their search results to a .BIB file for easy import to a reference manager and to download the full-text PDFs of preprints matching their search criteria.

Econometric analysis of multiple-input-multiple-output production technologies with ray-based input distance functions as suggested by Price and Henningsen (2023): "A Ray-Based Input Distance Function to Model Zero-Valued Output Quantities: Derivation and an Empirical Application", Journal of Productivity Analysis 60, p. 179-188, <doi:10.1007/s11123-023-00684-1>.

Calculate multiple statistics with confidence intervals for matched case-control data including risk difference, risk ratio, relative difference, and the odds ratio. Results are equivalent to those from Stata', and you can choose how to format your input data. Methods used are those described on page 56 the Stata documentation for "Epitab - Tables for Epidemologists" <https://www.stata.com/manuals/repitab.pdf>.