Building Regression and Classification Models

Consistent user interface to the most common regression and classification algorithms, such as random forest, neural networks, C5 trees and support vector machines, complemented with a handful of auxiliary functions, such as variable importance and a tuning function for the parameters.

Exact Variable-Subset Selection in Linear Regression

Exact and approximation algorithms for variable-subset selection in ordinary linear regression models. Either compute all submodels with the lowest residual sum of squares, or determine the single-best submodel according to a pre-determined statistical criterion. Hofmann et al. (2020) .

Regularized Linear Models

Algorithms compute robust estimators for loss functions in the concave convex (CC) family by the iteratively reweighted convex optimization (IRCO), an extension of the iteratively reweighted least squares (IRLS). The IRCO reduces the weight of the observation that leads to a large loss; it also provides weights to help identify outliers. Applications include robust (penalized) generalized linear models and robust support vector machines. The package also contains penalized Poisson, negative binomial, zero-inflated Poisson, zero-inflated negative binomial regression models and robust models with non-convex loss functions. Wang et al. (2014) , Wang et al. (2015) , Wang et al. (2016) , Wang (2021) , Wang (2024) .

Probability Distributions as S3 Objects

Tools to create and manipulate probability distributions using S3. Generics pdf(), cdf(), quantile(), and random() provide replacements for base R's d/p/q/r style functions. Functions and arguments have been named carefully to minimize confusion for students in intro stats courses. The documentation for each distribution contains detailed mathematical notes.

Model Wrappers for Poisson Regression

Bindings for Poisson regression models for use with the 'parsnip' package. Models include simple generalized linear models, Bayesian models, and zero-inflated Poisson models (Zeileis, Kleiber, and Jackman (2008) ).

Double/Debiased Machine Learning

Estimate common causal parameters using double/debiased machine learning as proposed by Chernozhukov et al. (2018) . 'ddml' simplifies estimation based on (short-)stacking as discussed in Ahrens et al. (2024) , which leverages multiple base learners to increase robustness to the underlying data generating process.

Breaks for Additive Season and Trend

Decomposition of time series into trend, seasonal, and remainder components with methods for detecting and characterizing abrupt changes within the trend and seasonal components. 'BFAST' can be used to analyze different types of satellite image time series and can be applied to other disciplines dealing with seasonal or non-seasonal time series, such as hydrology, climatology, and econometrics. The algorithm can be extended to label detected changes with information on the parameters of the fitted piecewise linear models. 'BFAST' monitoring functionality is described in Verbesselt et al. (2010) . 'BFAST monitor' provides functionality to detect disturbance in near real-time based on 'BFAST'- type models, and is described in Verbesselt et al. (2012) . 'BFAST Lite' approach is a flexible approach that handles missing data without interpolation, and will be described in an upcoming paper. Furthermore, different models can now be used to fit the time series data and detect structural changes (breaks).

A Tool for Processing and Analyzing Dendrometer Data

There are various functions for managing and cleaning data before the application of different approaches. This includes identifying and erasing sudden jumps in dendrometer data not related to environmental change, identifying the time gaps of recordings, and changing the temporal resolution of data to different frequencies. Furthermore, the package calculates daily statistics of dendrometer data, including the daily amplitude of tree growth. Various approaches can be applied to separate radial growth from daily cyclic shrinkage and expansion due to uptake and loss of stem water. In addition, it identifies periods of consecutive days with user-defined climatic conditions in daily meteorological data, then check what trees are doing during that period.

Automatic Generation of Exams in R for 'Sakai'

Automatic Generation of Exams in R for 'Sakai'. Question templates in the form of the 'exams' package (see < https://www.r-exams.org/>) are transformed into XML format required by 'Sakai'.

HMM-Based Model for Genotyping and Cross-Over Identification

Our method integrates information from all sequenced samples, thus avoiding loss of alleles due to low coverage. Moreover, it increases the statistical power to uncover sequencing or alignment errors .