This package provides robust parameter tuning and model training for predictive models applied across data sources where the data distribution varies slightly from source to source. This package implements three primary tuning methods: cross-validation-based internal tuning, external tuning, and the RobustTuneC method. External tuning includes a conservative option where parameters are tuned internally on the training data and validating on an external dataset, providing a slightly pessimistic estimate. It supports Lasso, Ridge, Random Forest, Boosting, and Support Vector Machine classifiers. Currently, only binary classification is supported. The response variable must be the first column of the dataset and a factor with exactly two levels. The tuning methods are based on the paper by Nicole Ellenbach, Anne-Laure Boulesteix, Bernd Bischl, Kristian Unger, and Roman Hornung (2021) "Improved Outcome Prediction Across Data Sources Through Robust Parameter Tuning" <doi:10.1007/s00357-020-09368-z>.

Define distribution families and fit them to interval-censored and interval-truncated data, where the truncation bounds may depend on the individual observation. The defined distributions feature density, probability, sampling and fitting methods as well as efficient implementations of the log-density log f(x) and log-probability log P(x0 <= X <= x1) for use in TensorFlow neural networks via the tensorflow package. Allows training parametric neural networks on interval-censored and interval-truncated data with flexible parameterization. Applications include Claims Development in Non-Life Insurance, e.g. modelling reporting delay distributions from incomplete data, see Bücher, Rosenstock (2022) <doi:10.1007/s13385-022-00314-4>.

This package provides a bagging predictor based on generalized linear models (GLMs) is implemented. The method is published in Song, Langfelder and Horvath (2013) <doi:10.1186/1471-2105-14-5>.

Enhances the R Optimization Infrastructure ('ROI') package with the DEoptim and DEoptimR package. DEoptim is used for unconstrained optimization and DEoptimR for constrained optimization.

The L-BFGS algorithm is a popular optimization algorithm for unconstrained optimization problems. Blaze is a high-performance C++ math library for dense and sparse arithmetic. This package provides a simple interface to the L-BFGS algorithm and allows users to optimize their objective functions with Blaze vectors and matrices in R and Rcpp'.

This package provides a S4 class has been created such that complex operations can be executed on each cell of a raster map. The raster of objects contains a raster map with the addition of a list of generic objects: one object for each raster cells. It allows to write few lines of R code for complex map algebra. Two environmental applications about frequency analysis of raster map of precipitation and creation of a raster map of soil water retention curves have been presented.

Estimates the rearranged dependence measure ('RDM') of two continuous random variables for different underlying measures. Furthermore, it provides a method to estimate the (SI)-rearrangement copula using empirical checkerboard copulas. It is based on the theoretical results presented in Strothmann et al. (2022) <arXiv:2201.03329> and Strothmann (2021) <doi:10.17877/DE290R-22733>.

Inspired by the classic RSA', we developed the improved Generalized Reporter Score-based Analysis (GRSA) method, implemented in the R package ReporterScore', along with comprehensive visualization methods and pathway databases. GRSA is a threshold-free method that works well with all types of biomedical features, such as genes, chemical compounds, and microbial species. Importantly, the GRSA supports multi-group and longitudinal experimental designs, because of the included multi-group-compatible statistical methods.

For the calculation of sample size or power in a two-group repeated measures design, accounting for attrition and accommodating a variety of correlation structures for the repeated measures; details of the method can be found in the scientific paper: Donald Hedeker, Robert D. Gibbons, Christine Waternaux (1999) <doi:10.3102/10769986024001070>.

R-based access to a large set of data variables relevant to forest ecology in British Columbia (BC), Canada. Layers are in raster format at 100m resolution in the BC Albers projection, hosted at the Federated Research Data Repository (FRDR) with <doi:10.20383/101.0283>. The collection includes: elevation; biogeoclimatic zone; wildfire; cutblocks; forest attributes from Hansen et al. (2013) <doi:10.1139/cjfr-2013-0401> and Beaudoin et al. (2017) <doi:10.1139/cjfr-2017-0184>; and rasterized Forest Insect and Disease Survey (FIDS) maps for a number of insect pest species, all covering the period 2001-2018. Users supply a polygon or point location in the province of BC, and rasterbc will download the overlapping raster tiles hosted at FRDR, merging them as needed and returning the result in R as a SpatRaster object. Metadata associated with these layers, and code for downloading them from their original sources can be found in the github repository <https://github.com/deankoch/rasterbc_src>.

This package provides access to the xylib C library for to import xy data from powder diffraction, spectroscopy and other experimental methods.

This framework aims to provide classes and methods for manipulating and processing of raster time series data (e.g. a time series of satellite images).

This package provides a simple and efficient way to read data from Paradox database files (.db) directly into R as modern tibble data frames. It uses the underlying pxlib C library, to handle the low-level file format details and provides a clean, user-friendly R interface.

This package provides a programmatic interface to the Web Service methods provided by the Global Biodiversity Information Facility (GBIF; <https://www.gbif.org/developer/summary>). GBIF is a database of species occurrence records from sources all over the globe. rgbif includes functions for searching for taxonomic names, retrieving information on data providers, getting species occurrence records, getting counts of occurrence records, and using the GBIF tile map service to make rasters summarizing huge amounts of data.

This package implements regression calibration methods for correcting measurement error in regression models using external or internal reliability studies. Methods are described in Carroll, Ruppert, Stefanski, and Crainiceanu (2006) "Measurement Error in Nonlinear Models: A Modern Perspective" <doi:10.1201/9781420010138>.

This package provides R-squared values and standardized regression coefficients for linear models applied to multiply imputed datasets as obtained by mice'. Confidence intervals, zero-order correlations, and alternative adjusted R-squared estimates are also available. The methods are described in Van Ginkel and Karch (2024) <doi:10.1111/bmsp.12344> and in Van Ginkel (2020) <doi:10.1007/s11336-020-09696-4>.

These functions are especially helpful when writing reports of data analysis using "Sweave".

Simplified scenario testing and sensitivity analysis, redesigned to use packages future and furrr'. Provides functions for generating function argument sets using one-factor-at-a-time (OFAT) and (sampled) permutations.

This function conducts variation partitioning and hierarchical partitioning to calculate the unique, shared (referred as to "common") and individual contributions of each predictor (or matrix) towards explained variation (R-square and adjusted R-square) on canonical analysis (RDA,CCA and db-RDA), applying the algorithm of Lai J.,Zou Y., Zhang J.,Peres-Neto P.(2022) Generalizing hierarchical and variation partitioning in multiple regression and canonical analyses using the rdacca.hp R package.Methods in Ecology and Evolution,13: 782-788 <DOI:10.1111/2041-210X.13800>.

Statistical tools based on the probabilistic properties of the record occurrence in a sequence of independent and identically distributed continuous random variables. In particular, tools to prepare a time series as well as distribution-free trend and change-point tests and graphical tools to study the record occurrence. Details about the implemented tools can be found in Castillo-Mateo et al. (2023a) <doi:10.18637/jss.v106.i05> and Castillo-Mateo et al. (2023b) <doi:10.1016/j.atmosres.2023.106934>.

An R interface for processing concentration-response datasets using Curvep, a response noise filtering algorithm. The algorithm was described in the publications (Sedykh A et al. (2011) <doi:10.1289/ehp.1002476> and Sedykh A (2016) <doi:10.1007/978-1-4939-6346-1_14>). Other parametric fitting approaches (e.g., Hill equation) are also adopted for ease of comparison. 3-parameter Hill equation from tcpl package (Filer D et al., <doi:10.1093/bioinformatics/btw680>) and 4-parameter Hill equation from Curve Class2 approach (Wang Y et al., <doi:10.2174/1875397301004010057>) are available. Also, methods for calculating the confidence interval around the activity metrics are also provided. The methods are based on the bootstrap approach to simulate the datasets (Hsieh J-H et al. <doi:10.1093/toxsci/kfy258>). The simulated datasets can be used to derive the baseline noise threshold in an assay endpoint. This threshold is critical in the toxicological studies to derive the point-of-departure (POD).

Features the multiple polynomial quadratic sieve (MPQS) algorithm for factoring large integers and a vectorized factoring function that returns the complete factorization of an integer. The MPQS is based off of the seminal work of Carl Pomerance (1984) <doi:10.1007/3-540-39757-4_17> along with the modification of multiple polynomials introduced by Peter Montgomery and J. Davis as outlined by Robert D. Silverman (1987) <doi:10.1090/S0025-5718-1987-0866119-8>. Utilizes the C library GMP (GNU Multiple Precision Arithmetic). For smaller integers, a simple Elliptic Curve algorithm is attempted followed by a constrained version of Pollard's rho algorithm. The Pollard's rho algorithm is the same algorithm used by the factorize function in the gmp package.

This package performs two-sample comparisons using the restricted mean survival time (RMST) when survival curves end at different time points between groups. This package implements a sensitivity approach that allows the threshold timepoint tau to be specified after the longest survival time in the shorter survival group. Two kinds of between-group contrast estimators (the difference in RMST and the ratio of RMST) are computed: Uno et al(2014)<doi:10.1200/JCO.2014.55.2208>, Uno et al(2022)<https://CRAN.R-project.org/package=survRM2>, Ueno and Morita(2023)<doi:10.1007/s43441-022-00484-z>.

Allows interaction with Interactive Brokers Trader Workstation <https://interactivebrokers.github.io/tws-api/>. Handles the connection over the network and the exchange of messages. Data is encoded and decoded between user and wire formats. Data structures and functionality closely mirror the official implementations.