This package provides a library of density, distribution function, quantile function, (bounded) raw moments and random generation for a collection of distributions relevant for the firm size literature. Additionally, the package contains tools to fit these distributions using maximum likelihood and evaluate these distributions based on (i) log-likelihood ratio and (ii) deviations between the empirical and parametrically implied moments of the distributions. We add flexibility by allowing the considered distributions to be combined into piecewise composite or finite mixture distributions, as well as to be used when truncated. See Dewitte (2020) <https://hdl.handle.net/1854/LU-8644700> for a description and application of methods available in this package.
This package defines a BigMatrix ReferenceClass which adds safety and convenience features to the filebacked.big.matrix class from the bigmemory package. BigMatrix protects against segfaults by monitoring and gracefully restoring the connection to on-disk data and it also protects against accidental data modification with a file-system-based permissions system. Utilities are provided for using BigMatrix-derived classes as assayData matrices within the Biobase package's eSet family of classes. BigMatrix provides some optimizations related to attaching to, and indexing into, file-backed matrices with dimnames. Additionally, the package provides a BigMatrixFactor class, a file-backed matrix with factor properties.
As different antipsychotic medications have different potencies, the doses of different medications cannot be directly compared. Various strategies are used to convert doses into a common reference so that comparison is meaningful. Chlorpromazine (CPZ) has historically been used as a reference medication into which other antipsychotic doses can be converted, as "chlorpromazine-equivalent doses". Using conversion keys generated from widely-cited scientific papers, e.g. Gardner et. al 2010 <doi:10.1176/appi.ajp.2009.09060802> and Leucht et al. 2016 <doi:10.1093/schbul/sbv167>, antipsychotic doses are converted to CPZ (or any specified antipsychotic) equivalents. The use of the package is described in the included vignette. Not for clinical use.
The Core Microbiome refers to the group of microorganisms that are consistently present in a particular environment, habitat, or host species. These microorganisms play a crucial role in the functioning and stability of that ecosystem. Identifying these microorganisms can contribute to the emerging field of personalized medicine. The CoreMicrobiomeR is designed to facilitate the identification, statistical testing, and visualization of this group of microorganisms.This package offers three key functions to analyze and visualize microbial community data. This package has been developed based on the research papers published by Pereira et al.(2018) <doi:10.1186/s12864-018-4637-6> and Beule L, Karlovsky P. (2020) <doi:10.7717/peerj.9593>.
The komacv-rg bundle provides packages that aid in creating CVs based on the komacv class and creating related documents, such as cover letters and cover sheets for job applications.
Concretely, the bundle consists of three packages: komacv-addons, komacv-lco, and komacv-multilang. komacv-addons is a small collection of add-ons and fixes for the komacv class; komacv-lco enables the use of letter class options from scrlttr2 also in komacv-based and other non-scrlttr2-based documents; komacv-multilang enables the provisioning of CVs in multiple languages and the selection of a language via Babel or Polyglossia.
This package provides a fast implementation with additional experimental features for testing, monitoring and dating structural changes in (linear) regression models. strucchangeRcpp features tests/methods from the generalized fluctuation test framework as well as from the F test (Chow test) framework. This includes methods to fit, plot and test fluctuation processes (e.g. cumulative/moving sum, recursive/moving estimates) and F statistics, respectively. These methods are described in Zeileis et al. (2002) <doi:10.18637/jss.v007.i02>. Finally, the breakpoints in regression models with structural changes can be estimated together with confidence intervals, and their magnitude as well as the model fit can be evaluated using a variety of statistical measures.
multiHiCcompare provides functions for joint normalization and difference detection in multiple Hi-C datasets. This extension of the original HiCcompare package now allows for Hi-C experiments with more than 2 groups and multiple samples per group. multiHiCcompare operates on processed Hi-C data in the form of sparse upper triangular matrices. It accepts four column (chromosome, region1, region2, IF) tab-separated text files storing chromatin interaction matrices. multiHiCcompare provides cyclic loess and fast loess (fastlo) methods adapted to jointly normalizing Hi-C data. Additionally, it provides a general linear model (GLM) framework adapting the edgeR package to detect differences in Hi-C data in a distance dependent manner.
Total Time on Test plot and routines for parameter estimation of any lifetime distribution implemented in R via maximum likelihood (ML) given a data set. It is implemented thinking on parametric survival analysis, but it feasible to use in parameter estimation of probability density or mass functions in any field. The main routines maxlogL and maxlogLreg are wrapper functions specifically developed for ML estimation. There are included optimization procedures such as nlminb and optim from base package, and DEoptim Mullen (2011) <doi:10.18637/jss.v040.i06>. Standard errors are estimated with numDeriv Gilbert (2011) <https://CRAN.R-project.org/package=numDeriv> or the option Hessian = TRUE of optim function.
This package provides functions for working with primary event censored distributions and Stan implementations for use in Bayesian modeling. Primary event censored distributions are useful for modeling delayed reporting scenarios in epidemiology and other fields (Charniga et al. (2024) <doi:10.48550/arXiv.2405.08841>). It also provides support for arbitrary delay distributions, a range of common primary distributions, and allows for truncation and secondary event censoring to be accounted for (Park et al. (2024) <doi:10.1101/2024.01.12.24301247>). A subset of common distributions also have analytical solutions implemented, allowing for faster computation. In addition, it provides multiple methods for fitting primary event censored distributions to data via optional dependencies.
This package provides implementations of the Diagnosis and Recommendation Integrated System (DRIS), the Modified DRIS (MDRIS), and the Plant Analysis with Standardized Scores (PASS) approaches for nutrient diagnosis in crops. These methods allow quantitative evaluation of nutrient imbalances using ratio-based indices and standardized scores, supporting improved fertilizer use efficiency and crop management decisions. The DRIS method is described in Walworth, J.L. and Sumner, M.E. (1987) <doi:10.1007/978-1-4612-4682-4_4>. The MDRIS approach is detailed in Beverly, R.B. (1987) <doi:10.1080/01904168709363672>. The PASS method combining DRIS and sufficiency ranges is presented in Baldock, J.O. and Schulte, E.E. (1996) <doi:10.2134/agronj1996.00021962008800030015x>.
The trigger strategy is a general framework for a multistage statistical design with multiple hypotheses, allowing an adaptive selection of interim analyses. The selection of interim stages can be associated with some prespecified endpoints which serve as the trigger. This selection allows us to refine the critical boundaries in hypotheses testing procedures, and potentially increase the statistical power. This package includes several trial designs using the trigger strategy. See Gou, J. (2023), "Trigger strategy in repeated tests on multiple hypotheses", Statistics in Biopharmaceutical Research, 15(1), 133-140, and Gou, J. (2022), "Sample size optimization and initial allocation of the significance levels in group sequential trials with multiple endpoints", Biometrical Journal, 64(2), 301-311.
This package implements statistical process control ('SPC') monitoring schemes for heterogeneous reliability observations using Accelerated Failure Time ('AFT') models integrated with continuous gamma frailty. It accommodates both uncensored and right-censored reliability observations in the presence of observed and unobserved covariates. Provides Phase I maximum likelihood estimation of Weibull AFT gamma frailty model parameters, and Phase II monitoring procedures including probability-limits-based control charts, exponentially weighted moving average ('EWMA') charts with conditional expected values, and likelihood-ratio cumulative sum ('CUSUM') control charts. Competing CUSUM schemes (ignoring unobserved frailty or both covariates) and Average Run Length ('ARL') simulation utilities are also provided. The statistical methodology is based on Asadzadeh (2022) <doi:10.1080/00949655.2021.1959582>.
git-remote-gcrypt is a Git remote helper to push and pull from repositories encrypted with GnuPG. It works with the standard Git transports, including repository hosting services like GitLab.
Remote helper programs are invoked by Git to handle network transport. This helper handles gcrypt: URLs that access a remote repository encrypted with GPG, using our custom format.
Supported locations are local, rsync:// and sftp://, where the repository is stored as a set of files, or instead any Git URL where gcrypt will store the same representation in a Git repository, bridged over arbitrary Git transport.
The aim is to provide confidential, authenticated Git storage and collaboration using typical untrusted file hosts or services.
This package provides a framework for scalable statistical computing on large on-disk matrices stored in HDF5 files. It provides efficient block-wise implementations of core linear-algebra operations (matrix multiplication, SVD, PCA, and QR decomposition) written in C++ and R, along with building blocks from which higher-level multivariate methods such as canonical correlation analysis can be constructed. These building blocks are designed not only for direct use, but also as foundational components for developing new statistical methods that must operate on datasets too large to fit in memory. The package supports data provided either as HDF5 files or standard R objects, and is intended for high-dimensional applications such as omics and precision-medicine research.
This package provides hardware-accelerated tools for performing rerandomization and randomization testing in experimental research. Using a JAX backend, the package enables exact rerandomization inference even for large experiments with hundreds of billions of possible randomizations. Key functionalities include generating pools of acceptable rerandomizations based on covariate balance, conducting exact randomization tests, and performing pre-analysis evaluations to determine optimal rerandomization acceptance thresholds. The package supports various hardware acceleration frameworks including CPU', CUDA', and METAL', making it versatile across accelerated computing environments. This allows researchers to efficiently implement stringent rerandomization designs and conduct valid inference even with large sample sizes. The package is partly based on Jerzak and Goldstein (2023) <doi:10.48550/arXiv.2310.00861>.
This package implements structural estimators to estimate preferences and correct for the sample selection bias of observed outcomes in matching markets. This includes one-sided matching of agents into groups (Klein, 2015) <doi:10.17863/CAM.5812> as well as two-sided matching of students to schools (Klein et al., 2024) <doi:10.1016/j.geb.2024.07.003>. The package also contains algorithms to find stable matchings in the three most common matching problems: the stable roommates problem (Irving, 1985) <doi:10.1016/0196-6774(85)90033-1>, the college admissions problem (Gale and Shapley, 1962) <doi:10.2307/2312726>, and the house allocation problem (Shapley and Scarf, 1974) <doi:10.1016/0304-4068(74)90033-0>.
Copula based Cox proportional hazards models for survival data subject to dependent censoring. This approach does not assume that the parameter defining the copula is known. The dependency parameter is estimated with other finite model parameters by maximizing a Pseudo likelihood function. The cumulative hazard function is estimated via estimating equations derived based on martingale ideas. Available copula functions include Frank, Gumbel and Normal copulas. Only Weibull and lognormal models are allowed for the censoring model, even though any parametric model that satisfies certain identifiability conditions could be used. Implemented methods are described in the article "Copula based Cox proportional hazards models for dependent censoring" by Deresa and Van Keilegom (2024) <doi:10.1080/01621459.2022.2161387>.
The extended neighbourhood rule for the k nearest neighbour ensemble where the neighbours are determined in k steps. Starting from the first nearest observation of the test point, the algorithm identifies a single observation that is closest to the observation at the previous step. At each base learner in the ensemble, this search is extended to k steps on a random bootstrap sample with a random subset of features selected from the feature space. The final predicted class of the test point is determined by using a majority vote in the predicted classes given by all base models. Amjad Ali, Muhammad Hamraz, Naz Gul, Dost Muhammad Khan, Saeed Aldahmani, Zardad Khan (2022) <doi:10.48550/arXiv.2205.15111>.
This package implements SelectBoost'-style variable selection workflows for functional data analysis. The package provides FDA-native design and preprocessing objects for raw curves, spline-basis expansions, Functional principal component analysis scores, and scalar covariates; grouped stability-selection routines based on repeated subject-level subsampling; multiple selector backends including lasso, group lasso, and sparse-group lasso; FDA-aware grouping functions and calibration helpers for SelectBoost'; method-comparison utilities; a formula interface; simulation, benchmarking, and validation helpers with mapped ground truth; targeted sensitivity-study utilities and shipped benchmark summaries for mean F1 comparisons between FDA-aware and plain SelectBoost workflows; small example datasets; and an optional adapter to the native stability-selection interface from the FDboost package.
readerwriterqueue provides a lock free queue supporting a two-thread use case (one consuming, one producing). Features include:
compatible with C++11 (supports moving objects instead of making copies)
fully generic (templated container of any type); just like
std::queue, you never need to allocate memory for elements yourselfallocates memory up front, in contiguous blocks
provides a
try_enqueuemethod which is guaranteed never to allocate memoryprovides an enqueue method which can dynamically grow the size of the queue as needed
Completely wait-free (no compare-and-swap loop). Enqueue and dequeue are always O(1) (not counting memory allocation)
On x86, the memory barriers compile down to no-ops
Fits up to 20 nonlinear lactation curve models to dairy animal milk yield data. Models fitted include exponential, polynomial, mixed logarithmic, inverse polynomial, and sigmoid families published between 1923 and 2000. Supports batch processing of multiple animals from a single CSV file, with flexible selection of animals and models. Produces per-animal parameter tables, goodness-of-fit metrics including R-squared (R2), Root Mean Square Error (RMSE), Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and a serial autocorrelation statistic, 15 diagnostic figures, and combined cross-animal comparison outputs. References: <doi:10.1085/jgp.5.4.441>, <doi:10.1038/216164a0>, <doi:10.1016/0301-6226(87)90003-0>, <doi:10.4141/cjas87-067>, <doi:10.3168/jds.S0022-0302(00)75136-8>.
Non-proportional hazard (NPH) is commonly observed in immuno-oncology studies, where the survival curves of the treatment and control groups show delayed separation. To properly account for NPH, several statistical methods have been developed. One such method is Max-Combo test, which is a straightforward and flexible hypothesis testing method that can simultaneously test for constant, early, middle, and late treatment effects. However, the majority of the Max-Combo test performed in clinical studies are unstratified, ignoring the important prognostic stratification factors. To fill this gap, we have developed an R package for stratified Max-Combo testing that accounts for stratified baseline factors. Our package explores various methods for calculating combined test statistics, estimating joint distributions, and determining the p-values.
This package provides tools to quantify how strongly pairs of words attract or repel each other in a text corpus, based on co-occurrence patterns. For each word pair, the phi coefficient (a correlation measure for binary variables) is computed from a document-term matrix and tested for significance, then classified as showing attraction (co-occurring more than chance would predict), repulsion (co-occurring less than chance would predict), or no significant relationship. A full pipeline is provided from raw text to a labeled network visualization. Unlike general-purpose pairwise correlation tools, wordorientation is built specifically for text: it handles tokenization and stopword removal, applies significance-based classification rather than reporting a raw correlation coefficient alone, and produces a ready-to-plot attraction/ repulsion network.
This package provides functions to perform the peer performance analysis of funds returns as described in Ardia and Boudt (2018) <doi:10.1016/j.jbankfin.2017.10.014>. For each fund, the package estimates the proportion of peers it outperforms, is equalled by, and is outperformed by, correcting for luck with the false discovery approach of Storey (2002) <doi:10.1111/1467-9868.00346>. Screenings can be based on factor-model alphas, Sharpe ratios, or modified Sharpe ratios, the latter using the equality test of Ardia and Boudt (2015) <doi:10.1016/j.frl.2015.02.008>. Funds can be screened within a universe or against a separate peer group, over rolling windows, and results come with bootstrap confidence intervals, summary, plot, and tidy data frame methods.