Non-imputational method for handling missing values in a prediction context, meaning that not only are there missing values in the training dataset, but also some values may be missing in future cases to be predicted. Based on the notion of regression averaging (Matloff (2017, ISBN: 9781498710916)).

This package provides functions implementing minimal distance estimation methods for parametric tail dependence models, as proposed in Einmahl, J.H.J., Kiriliouk, A., Krajina, A., and Segers, J. (2016) <doi:10.1111/rssb.12114> and Einmahl, J.H.J., Kiriliouk, A., and Segers, J. (2018) <doi:10.1007/s10687-017-0303-7>.

This package provides a unified tidyverse-compatible interface to R's machine learning packages. Wraps established implementations from glmnet', randomForest', xgboost', e1071', rpart', gbm', nnet', cluster', dbscan', and others - providing consistent function signatures, tidy tibble output, and unified ggplot2'-based visualization. The underlying algorithms are unchanged; tidylearn simply makes them easier to use together. Access raw model objects via the $fit slot for package-specific functionality. Methods include random forests Breiman (2001) <doi:10.1023/A:1010933404324>, LASSO regression Tibshirani (1996) <doi:10.1111/j.2517-6161.1996.tb02080.x>, elastic net Zou and Hastie (2005) <doi:10.1111/j.1467-9868.2005.00503.x>, support vector machines Cortes and Vapnik (1995) <doi:10.1007/BF00994018>, and gradient boosting Friedman (2001) <doi:10.1214/aos/1013203451>.

Interactive shiny application for working with textmining and text analytics. Various visualizations are provided.

Turn complex JSON data into tidy data frames.

Implementation of the tree-guided feature selection and logic aggregation approach introduced in Chen et al. (2024) <doi:10.1080/01621459.2024.2326621>. The method enables the selection and aggregation of large-scale rare binary features with a known hierarchical structure using a convex, linearly-constrained regularized regression framework. The package facilitates the application of this method to both linear regression and binary classification problems by solving the optimization problem via the smoothing proximal gradient descent algorithm (Chen et al. (2012) <doi:10.1214/11-AOAS514>).

Forecasting competitions are of increasing importance as a mean to learn best practices and gain knowledge. Data leakage is one of the most common issues that can often be found in competitions. Data leaks can happen when the training data contains information about the test data. For example: randomly chosen blocks of time series are concatenated to form a new time series, scale-shifts, repeating patterns in time series, white noise is added in the original time series to form a new time series, etc. tsdataleaks package can be used to detect data leakages in a collection of time series.

This package provides an intuitive interface for working with the competing risk endpoints. The package wraps the cmprsk package, and exports functions for univariate cumulative incidence estimates and competing risk regression. Methods follow those introduced in Fine and Gray (1999) <doi:10.1002/sim.7501>.

This package provides a set of vectorised functions to calculate medical equations used in transplantation, focused mainly on transplantation of abdominal organs. These functions include donor and recipient risk indices as used by NHS Blood & Transplant, OPTN/UNOS and Eurotransplant, tools for quantifying HLA mismatches, functions for calculating estimated glomerular filtration rate (eGFR), a function to calculate the APRI (AST to platelet ratio) score used in initial screening of suitability to receive a transplant from a hepatitis C seropositive donor and some biochemical unit converter functions. All functions are designed to work with either US or international units. References for the equations are provided in the vignettes and function documentation.

Runs tests using the testthat package but allows for multiple attempts for a single test. This is useful for noisy or flaky tests that generally pass but can fail due to occasional random errors, such as numeric instability or using random data.

Create a time-varying dataset using features, exposure, and look back specifications.

Generalization of the classification and regression tree (CART) model that partitions subjects into terminal nodes and tailors machine learning model to each terminal node.

Develop, evaluate, and score multiple choice examinations, psychological scales, questionnaires, and similar types of data involving sequences of choices among one or more sets of answers. This version of the package should be considered as brand new. Almost all of the functions have been changed, including their argument list. See the file NEWS.Rd in the Inst folder for more information. Using the package does not require any formal statistical knowledge beyond what would be provided by a first course in statistics in a social science department. There the user would encounter the concept of probability and how it is used to model data and make decisions, and would become familiar with basic mathematical and statistical notation. Most of the output is in graphical form.

This package provides three estimators for tensor response regression (TRR) and tensor predictor regression (TPR) models with tensor envelope structure. The three types of estimation approaches are generic and can be applied to any envelope estimation problems. The full Grassmannian (FG) optimization is often associated with likelihood-based estimation but requires heavy computation and good initialization; the one-directional optimization approaches (1D and ECD algorithms) are faster, stable and does not require carefully chosen initial values; the SIMPLS-type is motivated by the partial least squares regression and is computationally the least expensive. For details of TRR, see Li L, Zhang X (2017) <doi:10.1080/01621459.2016.1193022>. For details of TPR, see Zhang X, Li L (2017) <doi:10.1080/00401706.2016.1272495>. For details of 1D algorithm, see Cook RD, Zhang X (2016) <doi:10.1080/10618600.2015.1029577>. For details of ECD algorithm, see Cook RD, Zhang X (2018) <doi:10.5705/ss.202016.0037>. For more details of the package, see Zeng J, Wang W, Zhang X (2021) <doi:10.18637/jss.v099.i12>.

This package creates a table of descriptive statistics for factor and numeric columns in a data frame. Displays these by groups, if any. Highly customizable, with support for html and pdf provided by kableExtra'. Respects original column order, column labels, and factor level order. See ?tablet.data.frame and vignettes.

Fits 2D and 3D geometric transformations via Stan probabilistic programming engine ( Stan Development Team (2021) <https://mc-stan.org>). Returns posterior distribution for individual parameters of the fitted distribution. Allows for computation of LOO and WAIC information criteria (Vehtari A, Gelman A, Gabry J (2017) <doi:10.1007/s11222-016-9696-4>) as well as Bayesian R-squared (Gelman A, Goodrich B, Gabry J, and Vehtari A (2018) <doi:10.1080/00031305.2018.1549100>).

We present a range of simulations to aid researchers in determining appropriate sample sizes when performing critical thermal limits studies (e.g. CTmin/CTmin experiments). A number of wrapper functions are provided for plotting and summarising outputs from these simulations. This package is presented in van Steenderen, C.J.M., Sutton, G.F., Owen, C.A., Martin, G.D., and Coetzee, J.A. Sample size assessments for thermal physiology studies: An R package and R Shiny application. 2023. Physiological Entomology. <doi:10.1111/phen.12416>. The GUI version of this package is available on the R Shiny online server at: <https://clarkevansteenderen.shinyapps.io/ThermalSampleR_Shiny/> , or it is accessible via GitHub at <https://github.com/clarkevansteenderen/ThermalSampleR_Shiny/>. We would like to thank Grant Duffy (University of Otago, Dundedin, New Zealand) for granting us permission to use the source code for the Test of Total Equivalency function.

Density, distribution function, quantile function and random generation for the Truncated Generalised Gamma Distribution (also in log10(x) and ln(x) space).

Implementation of functions for fitting taper curves (a semiparametric linear mixed effects taper model) to diameter measurements along stems. Further functions are provided to estimate the uncertainty around the predicted curves, to calculate timber volume (also by sections) and marginal (e.g., upper) diameters. For cases where tree heights are not measured, methods for estimating additional variance in volume predictions resulting from uncertainties in tree height models (tariffs) are provided. The example data include the taper curve parameters for Norway spruce used in the 3rd German NFI fitted to 380 trees and a subset of section-wise diameter measurements of these trees. The functions implemented here are detailed in Kublin, E., Breidenbach, J., Kaendler, G. (2013) <doi:10.1007/s10342-013-0715-0>.

Algorithms for detecting population structure from the history of coalescent events recorded in phylogenetic trees. This method classifies each tip and internal node of a tree into disjoint sets characterized by similar coalescent patterns.

Binary ties limit the richness of network analyses as relations are unique. The two-mode structure contains a number of features lost when projection it to a one-mode network. Longitudinal datasets allow for an understanding of the causal relationship among ties, which is not the case in cross-sectional datasets as ties are dependent upon each other.

This package provides a problem solving environment (PSE) for fitting separable nonlinear models to measurements arising in physics and chemistry experiments, as described by Mullen & van Stokkum (2007) <doi:10.18637/jss.v018.i03> for its use in fitting time resolved spectroscopy data, and as described by Laptenok et al. (2007) <doi:10.18637/jss.v018.i08> for its use in fitting Fluorescence Lifetime Imaging Microscopy (FLIM) data, in the study of Förster Resonance Energy Transfer (FRET). `TIMP` also serves as the computation backend for the `GloTarAn` software, a graphical user interface for the package, as described in Snellenburg et al. (2012) <doi:10.18637/jss.v049.i03>.

Social Relation Model (SRM) analyses for single or multiple round-robin groups are performed. These analyses are either based on one manifest variable, one latent construct measured by two manifest variables, two manifest variables and their bivariate relations, or two latent constructs each measured by two manifest variables. Within-group t-tests for variance components and covariances are provided for single groups. For multiple groups two types of significance tests are provided: between-groups t-tests (as in SOREMO) and enhanced standard errors based on Lashley and Bond (1997) <DOI:10.1037/1082-989X.2.3.278>. Handling for missing values is provided.

Fitting time-varying coefficient models for single and multi-equation regressions, using kernel smoothing techniques.