# Matthias Schmid

#### 7 packages on CRAN

Implements boosting beta regression for potentially high-dimensional data (Mayr et al., 2018 <doi:10.1093/ije/dyy093>). The 'betaboost' package uses the same parametrization as 'betareg' (Cribari-Neto and Zeileis, 2010 <doi:10.18637/jss.v034.i02>) to make results directly comparable. The underlying algorithms are implemented via the R add-on packages 'mboost' (Hofner et al., 2014 <doi:10.1007/s00180-012-0382-5>) and 'gamboostLSS' (Mayr et al., 2012 <doi:10.1111/j.1467-9876.2011.01033.x>).

Provides data transformations, estimation utilities, predictive evaluation measures and simulation functions for discrete time survival analysis.

Building discrete-time survival trees and bagged trees based on the functionalities of the rpart package. Splitting criterion maximizes the likelihood of a covariate-free logistic discrete time hazard model.

Boosting models for fitting generalized additive models for location, shape and scale ('GAMLSS') to potentially high dimensional data.

Contains functions for estimation and model selection of kernel deep stacking networks.

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.

The package provides a variety of functions to estimate time-dependent true/false positive rates and AUC curves from a set of censored survival data.