Model-Based Boosting

Functional gradient descent algorithm (boosting) for optimizing general risk functions utilizing component-wise (penalised) least squares estimates or regression trees as base-learners for fitting generalized linear, additive and interaction models to potentially high-dimensional data. Models and algorithms are described in , a hands-on tutorial is available from . The package allows user-specified loss functions and base-learners.

Fast Gaussian Process Computation Using Vecchia's Approximation

Functions for fitting and doing predictions with Gaussian process models using Vecchia's (1988) approximation. Package also includes functions for reordering input locations, finding ordered nearest neighbors (with help from 'FNN' package), grouping operations, and conditional simulations. Covariance functions for spatial and spatial-temporal data on Euclidean domains and spheres are provided. The original approximation is due to Vecchia (1988) < http://www.jstor.org/stable/2345768>, and the reordering and grouping methods are from Guinness (2018) . Model fitting employs a Fisher scoring algorithm described in Guinness (2019) .

Optimally Robust Estimation

R infrastructure for optimally robust estimation in general smoothly parameterized models using S4 classes and methods as described Kohl, M., Ruckdeschel, P., and Rieder, H. (2010), , and in Rieder, H., Kohl, M., and Ruckdeschel, P. (2008), .

Optimally Robust Influence Curves and Estimators for Location and Scale

Functions for the determination of optimally robust influence curves and estimators in case of normal location and/or scale (see Chapter 8 in Kohl (2005) < https://epub.uni-bayreuth.de/839/2/DissMKohl.pdf>).

Interactive 'tourr' Using 'python'

Extends the functionality of the 'tourr' package by an interactive graphical user interface. The interactivity allows users to effortlessly refine their 'tourr' results by manual intervention, which allows for integration of expert knowledge and aids the interpretation of results. For more information on 'tourr' see Wickham et. al (2011) or < https://github.com/ggobi/tourr>.

Metabolomics Data Analysis Functions

A collection of functions for processing and analyzing metabolite data. The namesake function mrbin() converts 1D or 2D Nuclear Magnetic Resonance data into a matrix of values suitable for further data analysis and performs basic processing steps in a reproducible way. Negative values, a common issue in such data, can be replaced by positive values (). All used parameters are stored in a readable text file and can be restored from that file to enable exact reproduction of the data at a later time. The function fia() ranks features according to their impact on classifier models, especially artificial neural network models.

Statistical Inference of Vine Copulas

Provides tools for the statistical analysis of regular vine copula models, see Aas et al. (2009) and Dissman et al. (2013) . The package includes tools for parameter estimation, model selection, simulation, goodness-of-fit tests, and visualization. Tools for estimation, selection and exploratory data analysis of bivariate copula models are also provided.

Inferential Statistics

Computation of various confidence intervals (Altman et al. (2000), ISBN:978-0-727-91375-3; Hedderich and Sachs (2018), ISBN:978-3-662-56657-2) including bootstrapped versions (Davison and Hinkley (1997), ISBN:978-0-511-80284-3) as well as Hsu (Hedderich and Sachs (2018), ISBN:978-3-662-56657-2), permutation (Janssen (1997), ), bootstrap (Davison and Hinkley (1997), ISBN:978-0-511-80284-3), intersection-union (Sozu et al. (2015), ISBN:978-3-319-22005-5) and multiple imputation (Barnard and Rubin (1999), ) t-test; furthermore, computation of intersection-union z-test as well as multiple imputation Wilcoxon tests. Graphical visualization by volcano and Bland-Altman plots (Bland and Altman (1986), ; Shieh (2018), ).

Descriptive Statistics

Computation of standardized interquartile range (IQR), Huber-type skipped mean (Hampel (1985), ), robust coefficient of variation (CV) (Arachchige et al. (2019), ), robust signal to noise ratio (SNR), z-score, standardized mean difference (SMD), as well as functions that support graphical visualization such as boxplots based on quartiles (not hinges), negative logarithms and generalized logarithms for 'ggplot2' (Wickham (2016), ISBN:978-3-319-24277-4).

Tools for Statistical Disclosure Control in Research Data Centers

Tools for researchers to explicitly show that their results comply to rules for statistical disclosure control imposed by research data centers. These tools help in checking descriptive statistics and models and in calculating extreme values that are not individual data. Also included is a simple function to create log files. The methods used here are described in the "Guidelines for the checking of output based on microdata research" by Bond, Brandt, and de Wolf (2015) < https://ec.europa.eu/eurostat/cros/system/files/dwb_standalone-document_output-checking-guidelines.pdf>.