Find functions in an unstructured directory and explore their dependencies. Sourcing of R source files is performed without side-effects: from R scripts that have executable code and function definitions only functions are sourced.

Kiener distributions K1, K2, K3, K4 and K7 to characterize distributions with left and right, symmetric or asymmetric fat tails in finance, neuroscience and other disciplines. Two algorithms to estimate the distribution parameters, quantiles, value-at-risk and expected shortfall. IMPORTANT: Standardization has been changed in versions >= 2.0.0 to get sd = 1 when kappa = Inf rather than 2*pi/sqrt(3) in versions <= 1.8.6. This affects parameter g (other parameters stay unchanged). Do not update if you need consistent comparisons with previous results for the g parameter.

This package implements fast, scalable optimization algorithms for fitting generalized principal components analysis (GLM-PCA) models, as described in "A Generalization of Principal Components Analysis to the Exponential Family" Collins M, Dasgupta S, Schapire RE (2002, ISBN:9780262271738), and subsequently "Feature Selection and Dimension Reduction for Single-Cell RNA-Seq Based on a Multinomial Model" Townes FW, Hicks SC, Aryee MJ, Irizarry RA (2019) <doi:10.1186/s13059-019-1861-6>.

This package provides a comprehensive set of datasets and tools for causal inference research. The package includes data from clinical trials, cancer studies, epidemiological surveys, environmental exposures, and health-related observational studies. Designed to facilitate causal analysis, risk assessment, and advanced statistical modeling, it leverages datasets from packages such as causalOT', survival', causalPAF', evident', melt', and sanon'. The package is inspired by the foundational work of Pearl (2009) <doi:10.1017/CBO9780511803161> on causal inference frameworks.

This package provides a tool to create hydroclimate scenarios, stress test systems and visualize system performance in scenario-neutral climate change impact assessments. Scenario-neutral approaches stress-test the performance of a modelled system by applying a wide range of plausible hydroclimate conditions (see Brown & Wilby (2012) <doi:10.1029/2012EO410001> and Prudhomme et al. (2010) <doi:10.1016/j.jhydrol.2010.06.043>). These approaches allow the identification of hydroclimatic variables that affect the vulnerability of a system to hydroclimate variation and change. This tool enables the generation of perturbed time series using a range of approaches including simple scaling of observed time series (e.g. Culley et al. (2016) <doi:10.1002/2015WR018253>) and stochastic simulation of perturbed time series via an inverse approach (see Guo et al. (2018) <doi:10.1016/j.jhydrol.2016.03.025>). It incorporates Richardson-type weather generator model configurations documented in Richardson (1981) <doi:10.1029/WR017i001p00182>, Richardson and Wright (1984), as well as latent variable type model configurations documented in Bennett et al. (2018) <doi:10.1016/j.jhydrol.2016.12.043>, Rasmussen (2013) <doi:10.1002/wrcr.20164>, Bennett et al. (2019) <doi:10.5194/hess-23-4783-2019> to generate hydroclimate variables on a daily basis (e.g. precipitation, temperature, potential evapotranspiration) and allows a variety of different hydroclimate variable properties, herein called attributes, to be perturbed. Options are included for the easy integration of existing system models both internally in R and externally for seamless stress-testing'. A suite of visualization options for the results of a scenario-neutral analysis (e.g. plotting performance spaces and overlaying climate projection information) are also included. Version 1.0 of this package is described in Bennett et al. (2021) <doi:10.1016/j.envsoft.2021.104999>. As further developments in scenario-neutral approaches occur the tool will be updated to incorporate these advances.

This package performs fast detection of interactions in large-scale data using the method of random intersection trees introduced in Shah, R. D. and Meinshausen, N. (2014) <http://www.jmlr.org/papers/v15/shah14a.html>. The algorithm finds potentially high-order interactions in high-dimensional binary two-class classification data, without requiring lower order interactions to be informative. The search is particularly fast when the matrices of predictors are sparse. It can also be used to perform market basket analysis when supplied with a single binary data matrix. Here it will find collections of columns which for many rows contain all 1's.

The aim of the package is to provide some basic functions for doing statistics with trapezoidal fuzzy numbers. In particular, the package contains several functions for simulating trapezoidal fuzzy numbers, as well as for calculating some central tendency measures (mean and two types of median), some scale measures (variance, ADD, MDD, Sn, Qn, Tn and some M-estimators) and one diversity index and one inequality index. Moreover, functions for calculating the 1-norm distance, the mid/spr distance and the (phi,theta)-wabl/ldev/rdev distance between fuzzy numbers are included, and a function to calculate the value phi-wabl given a sample of trapezoidal fuzzy numbers.

This package provides functions to implement the Flexible cFDR (Hutchinson et al. (2021) <doi:10.1371/journal.pgen.1009853>) and Binary cFDR (Hutchinson et al. (2021) <doi:10.1101/2021.10.21.465274>) methodologies to leverage auxiliary data from arbitrary distributions, for example functional genomic data, with GWAS p-values to generate re-weighted p-values.

Lightweight utilities to estimate autoregressive (AR) and autoregressive moving average (ARMA) noise models from residuals and apply matched generalized least squares to whiten functional magnetic resonance imaging (fMRI) design and data matrices. The ARMA estimator follows a classic 1982 approach <doi:10.1093/biomet/69.1.81>, and a restricted AR family mirrors workflows described by Cox (2012) <doi:10.1016/j.neuroimage.2011.08.056>.

Fresh biomass determination is the key to evaluating crop genotypes response to diverse input and stress conditions and forms the basis for calculating net primary production. However, as conventional phenotyping approaches for measuring fresh biomass is time-consuming, laborious and destructive, image-based phenotyping methods are being widely used now. In the image-based approach, the fresh weight of the above-ground part of the plant depends on the projected area. For determining the projected area, the visual image of the plant is converted into the grayscale image by simply averaging the Red(R), Green (G) and Blue (B) pixel values. Grayscale image is then converted into a binary image using Otsuâ s thresholding method Otsu, N. (1979) <doi:10.1109/TSMC.1979.4310076> to separate plant area from the background (image segmentation). The segmentation process was accomplished by selecting the pixels with values over the threshold value belonging to the plant region and other pixels to the background region. The resulting binary image consists of white and black pixels representing the plant and background regions. Finally, the number of pixels inside the plant region was counted and converted to square centimetres (cm2) using the reference object (any object whose actual area is known previously) to get the projected area. After that, the projected area is used as input to the machine learning model (Linear Model, Artificial Neural Network, and Support Vector Regression) to determine the plant's fresh weight.

This is a method for Allele-specific DNA Copy Number Profiling using Next-Generation Sequencing. Given the allele-specific coverage at the variant loci, this program segments the genome into regions of homogeneous allele-specific copy number. It requires, as input, the read counts for each variant allele in a pair of case and control samples. For detection of somatic mutations, the case and control samples can be the tumor and normal sample from the same individual.

Implementation of the Factorized Binary Search (FaBiSearch) methodology for the estimation of the number and the location of multiple change points in the network (or clustering) structure of multivariate high-dimensional time series. The method is motivated by the detection of change points in functional connectivity networks for functional magnetic resonance imaging (fMRI) data. FaBiSearch uses non-negative matrix factorization (NMF), an unsupervised dimension reduction technique, and a new binary search algorithm to identify multiple change points. It requires minimal assumptions. Lastly, we provide interactive, 3-dimensional, brain-specific network visualization capability in a flexible, stand-alone function. This function can be conveniently used with any node coordinate atlas, and nodes can be color coded according to community membership, if applicable. The output is an elegantly displayed network laid over a cortical surface, which can be rotated in the 3-dimensional space. The main routines of the package are detect.cps(), for multiple change point detection, est.net(), for estimating a network between stationary multivariate time series, net.3dplot(), for plotting the estimated functional connectivity networks, and opt.rank(), for finding the optimal rank in NMF for a given data set. The functions have been extensively tested on simulated multivariate high-dimensional time series data and fMRI data. For details on the FaBiSearch methodology, please see Ondrus et al. (2021) <arXiv:2103.06347>. For a more detailed explanation and applied examples of the fabisearch package, please see Ondrus and Cribben (2022), preprint.

This package provides functions to fit regression models for bounded continuous and discrete responses. In case of bounded continuous responses (e.g., proportions and rates), available models are the flexible beta (Migliorati, S., Di Brisco, A. M., Ongaro, A. (2018) <doi:10.1214/17-BA1079>), the variance-inflated beta (Di Brisco, A. M., Migliorati, S., Ongaro, A. (2020) <doi:10.1177/1471082X18821213>), the beta (Ferrari, S.L.P., Cribari-Neto, F. (2004) <doi:10.1080/0266476042000214501>), and their augmented versions to handle the presence of zero/one values (Di Brisco, A. M., Migliorati, S. (2020) <doi:10.1002/sim.8406>) are implemented. In case of bounded discrete responses (e.g., bounded counts, such as the number of successes in n trials), available models are the flexible beta-binomial (Ascari, R., Migliorati, S. (2021) <doi:10.1002/sim.9005>), the beta-binomial, and the binomial are implemented. Inference is dealt with a Bayesian approach based on the Hamiltonian Monte Carlo (HMC) algorithm (Gelman, A., Carlin, J. B., Stern, H. S., Rubin, D. B. (2014) <doi:10.1201/b16018>). Besides, functions to compute residuals, posterior predictives, goodness of fit measures, convergence diagnostics, and graphical representations are provided.

This package provides a collection of methods for modeling time-to-event data using both functional and scalar predictors. It implements functional data analysis techniques for estimation and inference, allowing researchers to assess the impact of functional covariates on survival outcomes, including time-to-single event and recurrent event outcomes.

This package provides a simple and efficient wrapper around the fastest Fourier transform in the west (FFTW) library <http://www.fftw.org/>.

Easily use Font Awesome icons as shiny favicons (the icons that appear on browser tabs). Font Awesome (<https://fontawesome.com/>) is a popular set of icons that can be used in web pages. favawesome provides a simple way to use these icons as favicons in shiny applications and other HTML pages.

Tests for Kaiser-Meyer-Olkin (KMO) and communalities in a dataset. It provides a final sample by removing variables in a iterable manner while keeping account of the variables that were removed in each step. It follows the best practices and assumptions according to Hair, Black, Babin & Anderson (2018, ISBN:9781473756540).

Manages a file system cache. Regular files can be moved or copied to the cache folder. Sub-folders can be created in order to organize the files. Files can be located inside the cache using a glob function. Text contents can be easily stored in and retrieved from the cache using dedicated functions. It can be used for an application or a package, as a global cache, or as a per-user cache, in which case the standard OS user cache folder will be used (e.g.: on Linux $HOME/.cache/R/my_app_or_pkg_cache_folder).

Filling in the missing entries of a partially observed data is one of fundamental problems in various disciplines of mathematical science. For many cases, data at our interests have canonical form of matrix in that the problem is posed upon a matrix with missing values to fill in the entries under preset assumptions and models. We provide a collection of methods from multiple disciplines under Matrix Completion, Imputation, and Inpainting. See Davenport and Romberg (2016) <doi:10.1109/JSTSP.2016.2539100> for an overview of the topic.

Computes factorial A-, D- and E-optimal designs for two-colour cDNA microarray experiments.

The heterogeneous treatment effect estimation procedure proposed by Imai and Ratkovic (2013)<DOI: 10.1214/12-AOAS593>. The proposed method is applicable, for example, when selecting a small number of most (or least) efficacious treatments from a large number of alternative treatments as well as when identifying subsets of the population who benefit (or are harmed by) a treatment of interest. The method adapts the Support Vector Machine classifier by placing separate LASSO constraints over the pre-treatment parameters and causal heterogeneity parameters of interest. This allows for the qualitative distinction between causal and other parameters, thereby making the variable selection suitable for the exploration of causal heterogeneity. The package also contains a class of functions, CausalANOVA, which estimates the average marginal interaction effects (AMIEs) by a regularized ANOVA as proposed by Egami and Imai (2019). It contains a variety of regularization techniques to facilitate analysis of large factorial experiments.

This package provides the probability density function (PDF), cumulative distribution function (CDF), the first-order and second-order partial derivatives of the PDF, and a fitting function for the diffusion decision model (DDM; e.g., Ratcliff & McKoon, 2008, <doi:10.1162/neco.2008.12-06-420>) with across-trial variability in the drift rate. Because the PDF, its partial derivatives, and the CDF of the DDM both contain an infinite sum, they need to be approximated. fddm implements all published approximations (Navarro & Fuss, 2009, <doi:10.1016/j.jmp.2009.02.003>; Gondan, Blurton, & Kesselmeier, 2014, <doi:10.1016/j.jmp.2014.05.002>; Blurton, Kesselmeier, & Gondan, 2017, <doi:10.1016/j.jmp.2016.11.003>; Hartmann & Klauer, 2021, <doi:10.1016/j.jmp.2021.102550>) plus new approximations. All approximations are implemented purely in C++ providing faster speed than existing packages.

Measure fairness metrics in one place for many models. Check how big is model's bias towards different races, sex, nationalities etc. Use measures such as Statistical Parity, Equal odds to detect the discrimination against unprivileged groups. Visualize the bias using heatmap, radar plot, biplot, bar chart (and more!). There are various pre-processing and post-processing bias mitigation algorithms implemented. Package also supports calculating fairness metrics for regression models. Find more details in (WiÅ niewski, Biecek (2021)) <arXiv:2104.00507>.

Easily create graphs of the inter-relationships between functions in an environment.