This package provides a set of functions for applying a restricted linear algebra to the analysis of count-based data. See the accompanying preprint manuscript: "Normalizing need not be the norm: count-based math for analyzing single-cell data" Church et al (2022) <doi:10.1101/2022.06.01.494334> This tool is specifically designed to analyze count matrices from single cell RNA sequencing assays. The tools implement several count-based approaches for standard steps in single-cell RNA-seq analysis, including scoring genes and cells, comparing cells and clustering, calculating differential gene expression, and several methods for rank reduction. There are many opportunities for further optimization that may prove useful in the analysis of other data. We provide the source code freely available at <https://github.com/shchurch/countland> and encourage users and developers to fork the code for their own purposes.

Supervised learning from a source distribution (with known segmentation into cell sub-populations) to fit a target distribution with unknown segmentation. It relies regularized optimal transport to directly estimate the different cell population proportions from a biological sample characterized with flow cytometry measurements. It is based on the regularized Wasserstein metric to compare cytometry measurements from different samples, thus accounting for possible mis-alignment of a given cell population across sample (due to technical variability from the technology of measurements). Supervised learning technique based on the Wasserstein metric that is used to estimate an optimal re-weighting of class proportions in a mixture model Details are presented in Freulon P, Bigot J and Hejblum BP (2023) <doi:10.1214/22-AOAS1660>.

Reads Word documents containing incomplete bibliographic references and produces an updated file with standardized and complete references. The package provides functions to retrieve missing authors, titles, journal details, volume, issue, and page numbers. Digital object identifiers (DOIs) are retrieved using the CrossRef application programming interface (API) <https://api.crossref.org>, and references are formatted following DOI-based citation standards as described by Paskin (2010) <doi:10.1000/182> and the citation.doi.org service <https://citation.doi.org>. The package is intended to simplify reference preparation for scientific journal submissions.

Data sets used for copula modeling in addition to those in the R package copula'. These include a random subsample from the US National Education Longitudinal Study (NELS) of 1988 and nursing home data from Wisconsin.

This package provides a Shiny application to conduct comprehensive analysis of latent means including the examination of group equivalency, propensity score analysis, measurement invariance analysis, and assessment of latent mean differences of equivalent groups with invariant data. Group equivalency and propensity score analyses are implemented using the MatchIt package [Ho et al. (2011) <doi:10.18637/jss.v042.i08>], ensuring robust control for covariates. Structural equation modeling and invariance testing rely heavily on the lavaan package [Rosseel (2012) <doi:10.18637/jss.v048.i02>], providing a flexible and powerful modeling framework. The application also integrates modified functions from Hammack-Brown et al. (2021) <doi:10.1002/hrdq.21452> to support factor ratio testing and the list-and-delete procedure.

The Genetic Algorithm (GA) is used to perform changepoint analysis in time series data. The package also includes an extended island version of GA, as described in Lu, Lund, and Lee (2010, <doi:10.1214/09-AOAS289>). By mimicking the principles of natural selection and evolution, GA provides a powerful stochastic search technique for solving combinatorial optimization problems. In changepointGA', each chromosome represents a changepoint configuration, including the number and locations of changepoints, hyperparameters, and model parameters. The package employs genetic operatorsâ selection, crossover, and mutationâ to iteratively improve solutions based on the given fitness (objective) function. Key features of changepointGA include encoding changepoint configurations in an integer format, enabling dynamic and simultaneous estimation of model hyperparameters, changepoint configurations, and associated parameters. The detailed algorithmic implementation can be found in the package vignettes and in the paper of Li (2024, <doi:10.48550/arXiv.2410.15571>).

The ConNEcT approach investigates the pairwise association strength of binary time series by calculating contingency measures and depicts the results in a network. The package includes features to explore and visualize the data. To calculate the pairwise concurrent or temporal sequenced relationship between the variables, the package provides seven contingency measures (proportion of agreement, classical & corrected Jaccard, Cohen's kappa, phi correlation coefficient, odds ratio, and log odds ratio), however, others can easily be implemented. The package also includes non-parametric significance tests, that can be applied to test whether the contingency value quantifying the relationship between the variables is significantly higher than chance level. Most importantly this test accounts for auto-dependence and relative frequency.See Bodner et al.(2021) <doi: 10.1111/bmsp.12222>.Finally, a network can be drawn. Variables depicted the nodes of the network, with the node size adapted to the prevalence. The association strength between the variables defines the undirected (concurrent) or directed (temporal sequenced) links between the nodes. The results of the non-parametric significance test can be included by depicting either all links or only the significant ones. Tutorial see Bodner et al.(2021) <doi:10.3758/s13428-021-01760-w>.

Allows clustering of incomplete observations by addressing missing values using multiple imputation. For achieving this goal, the methodology consists in three steps, following Audigier and Niang 2022 <doi:10.1007/s11634-022-00519-1>. I) Missing data imputation using dedicated models. Four multiple imputation methods are proposed, two are based on joint modelling and two are fully sequential methods, as discussed in Audigier et al. (2021) <doi:10.48550/arXiv.2106.04424>. II) cluster analysis of imputed data sets. Six clustering methods are available (distances-based or model-based), but custom methods can also be easily used. III) Partition pooling. The set of partitions is aggregated using Non-negative Matrix Factorization based method. An associated instability measure is computed by bootstrap (see Fang, Y. and Wang, J., 2012 <doi:10.1016/j.csda.2011.09.003>). Among applications, this instability measure can be used to choose a number of clusters with missing values. The package also proposes several diagnostic tools to tune the number of imputed data sets, to tune the number of iterations in fully sequential imputation, to check the fit of imputation models, etc.

Cure dependent censoring regression models for long-term survival multivariate data. These models are based on extensions of the frailty models, capable to accommodating the cure fraction and the dependence between failure and censoring times, with Weibull and piecewise exponential marginal distributions. Theoretical details regarding the models implemented in the package can be found in Schneider et al. (2022) <doi:10.1007/s10651-022-00549-0>.

The COVID Symptom Study is a non-commercial project that uses a free mobile app to facilitate real-time data collection of symptoms, exposures, and risk factors related to COVID19. The package allows easy access to summary statistics data from COVID Symptom Study Sweden.

Provide step by step guided tours of Shiny applications.

Wrapper of .Call() that runs exit handlers to clean up C resources. Helps managing C (non-R) resources while using the R API.

Balance sheet and income statement metrics, investment analysis methods, valuation methods, loan amortization schedules, and Capital Asset Pricing Model.

This package implements cointegration tests with structural breaks designed for small sample sizes, following the methodology of Trinh (2022) <https://ideas.repec.org/p/ema/worpap/2022-01.html>. Supports models with no breaks, breaks in constant only, and breaks in both constant and slope. Provides endogenous break date detection using ADF or SSR minimization criteria, with small-sample adjusted critical values via response surface methodology.

This package provides a set of functions to fit a boosting conditional logit model.

Generate and analyse crossover designs from combinatorial or search algorithms as well as from literature and a GUI to access them.

Visualize the connectedness of factors in two-way tables. Perform two-way filtering to improve the degree of connectedness. See Weeks & Williams (1964) <doi:10.1080/00401706.1964.10490188>.

It provides functions that calculate Mahalanobis distance, Euclidean distance, Manhattan distance, Chebyshev distance, Hamming distance, Canberra distance, Minkowski dissimilarity (distance defined for p >= 1), Cosine dissimilarity, Bhattacharyya dissimilarity, Jaccard distance, Hellinger distance, Bray-Curtis dissimilarity, Sorensen-Dice dissimilarity between each pair of species in a list of data frames. These statistics are fundamental in various fields, such as cluster analysis, classification, and other applications of machine learning and data mining, where assessing similarity or dissimilarity between data is crucial. The package is designed to be flexible and easily integrated into data analysis workflows, providing reliable tools for evaluating distances in multidimensional contexts.

Conditioned Latin hypercube sampling, as published by Minasny and McBratney (2006) <DOI:10.1016/j.cageo.2005.12.009>. This method proposes to stratify sampling in presence of ancillary data. An extension of this method, which propose to associate a cost to each individual and take it into account during the optimisation process, is also proposed (Roudier et al., 2012, <DOI:10.1201/b12728>).

Providing a cluster allocation for n samples, either with an $n \times p$ data matrix or an $n \times n$ distance matrix, a bootstrap procedure is performed. The proportion of bootstrap replicates where a pair of samples cluster in the same cluster indicates who tightly the samples in a particular cluster clusters together.

This is a one-function package that will pass only unique values to a computationally-expensive function that returns an output of the same length as the input. In importing and working with tidy data, it is common to have index columns, often including time stamps that are far from unique. Some functions to work with these such as text conversion to other variable types (e.g. as.POSIXct()), various grep()-based functions, and often the cut() function are relatively slow when working with tens of millions of rows or more.

This package contains functions which can be used to calculate Pesticide Risk Metric values in aquatic environments from concentrations of multiple pesticides with known species sensitive distributions (SSDs). Pesticides provided by this package have all be validated however if the user has their own pesticides with SSD values they can append them to the pesticide_info table to include them in estimates.

This package provides simple, reproducible access to datasets from the Chicago Open Data portal <https://data.cityofchicago.org/>. Functions return results as tidy tibbles and support optional filtering, sorting, and row limits via the Socrata API.

Tests convergence in macro-financial panels combining Dynamic Factor Models (DFM) and mean-reverting, discrete-time Ornstein-Uhlenbeck/AR(1) factor processes. Provides: (i) static factor extraction with VAR stability checks, Portmanteau tests and rolling out-of-sample R^2, in the spirit of Stock and Watson (2002) <doi:10.1198/073500102317351921> and the Generalized Dynamic Factor Model of Forni, Hallin, Lippi and Reichlin (2000) <doi:10.1162/003465300559037>; (ii) cointegration analysis a la Johansen (1988) <doi:10.1016/0165-1889(88)90041-3>; (iii) Bayesian factor-OU/AR(1) estimation with convergence and half-life summaries grounded in Uhlenbeck and Ornstein (1930) <doi:10.1103/PhysRev.36.823> and Vasicek (1977) <doi:10.1016/0304-405X(77)90016-2>, with full Markov chain Monte Carlo convergence diagnostics; (iv) heteroskedasticity-consistent (HC) and, when the suggested sandwich (Zeileis (2004) <doi:10.18637/jss.v011.i10>) and lmtest packages are available, heteroskedasticity- and autocorrelation- consistent (HAC) robust inference, with a self-contained HC fallback; (v) coupling significance tests based on time-shift / block-bootstrap nulls that preserve marginal dynamics while breaking cross-series dependence; and (vi) optional PLS-based factor preselection (Mevik and Wehrens (2007) <doi:10.18637/jss.v018.i02>). Functions emphasize reproducibility (explicit seeds throughout) and clear, publication-ready summaries.