Testing the additional predictive value of high-dimensional molecular data (2009)
Boulesteix, Anne-Laure, Hothorn, Torsten
While high-dimensional molecular data such as microarray gene expression data have been used for disease outcome prediction or diagnosis purposes for about ten years in biomedical research, the...
Fenske, Nora, Kneib, Thomas, Hothorn, Torsten
Ordinary linear and generalized linear regression models relate the mean of a response variable to a linear combination of covariate effects and, as a consequence, focus on average properties of the...
Hothorn, Ludwig A, Vaeth, Michael, Hothorn, Torsten
Abstract One possibility for the statistical evaluation of trends in epidemiological exposure studies is the use of a trend test for data organized in a 2 × k contingency table. Commonly, the...
Variable Selection and Model Choice in Geoadditive Regression Models (2009)
Thomas Kneib, Torsten Hothorn, Gerhard Tutz, Thomas Kneib, Torsten Hothorn, Gerhard Tutz
Model choice and variable selection are issues of major concern in practi-cal regression analyses. We propose a boosting procedure that facilitates both tasks in a class of complex geoadditive...
References Bias in Random Forest Variable Importance Measures (2009)
Carolin Strobl, Anne-laure Boulesteix, Achim Zeileis, Torsten Hothorn, J. Friedman, R. Olshen, ...
Eerdewegh (2005). Identifying SNPs predictive of phenotype using random
Spatial Smoothing Techniques for the Assessment of Habitat Suitability (2008)
Thomas Kneib, Jörg Müller, Nationalparkverwaltung Bayerischer Wald, Torsten Hothorn
Precise knowledge about factors influencing the habitat suitability of a certain species forms the basis for the implementation of effective programs to conserve biological diversity. Such knowledge...
Variable Selection and Model Choice in Structured Survival Models (2008)
Hofner, Benjamin, Hothorn, Torsten, Kneib, Thomas
In many situations, medical applications ask for flexible survival models that allow to extend the classical Cox-model via the inclusion of time-varying and nonparametric effects. These structured...
Schmid, Matthias, Potapov, Sergej, Pfahlberg, Annette, Hothorn, Torsten
Boosting is one of the most important methods for fitting regression models and building prediction rules from high-dimensional data. A notable feature of boosting is that the technique has a...
Maximally Selected Rank Statistics in R (2008)
Torsten Hothorn, Berthold Lausen
This document gives some examples on how to use the maxstat package and is basically an extention to Hothorn and Lausen (2002). 1
Flexible boosting of accelerated failure time models (2008)
Schmid, Matthias, Hothorn, Torsten
Abstract Background When boosting algorithms are used for building survival models from high-dimensional data, it is common to fit a Cox proportional hazards model or to use least squares techniques...
Exploratory and Inferential Analysis of Benchmark Experiments (2008)
Eugster, Manuel J. A., Hothorn, Torsten, Leisch, Friedrich
Benchmark experiments produce data in a very specific format. The observations are drawn from the performance distributions of the candidate algorithms on resampled data sets. In this paper we...
Boosting Algorithms: Regularization, Prediction and Model Fitting (2008)
Bühlmann, Peter, Hothorn, Torsten
We present a statistical perspective on boosting. Special emphasis is given to estimating potentially complex parametric or nonparametric models, including generalized linear and additive models as...
Rejoinder: Boosting Algorithms: Regularization, Prediction and Model Fitting (2008)
Bühlmann, Peter, Hothorn, Torsten
Rejoinder to ``Boosting Algorithms: Regularization, Prediction and Model Fitting'' [arXiv:0804.2752]
Bundling in R 2003/03/20 Bundling Predictors in R (2008)
Constructing a good classifier is a three step procedure: 1. learning of different classifiers C 1,..., C K (K is rather large in R), 2. selecting the ”optimal ” C k (with minimum estimated...
Torsten Hothorn, Berthold Lausen
tree classifiers for laser scanning images: a data- and simulation-based strategy
Torsten Hothorn, Peter Bühlmann
This document reproduces the data analyses presented in Bühlmann and Hothorn (2008). For a description of the theory behind applications shown here we refer to the original manuscript. Note: The...
Bundling Predictors 2004/11/17 Predictions (2008)
We assume that the distribution of a response variable y depends on a vector of input variables (or covariates) x through a function f: Special cases are: D(y|x) = D(y|f(x)). y = f(x) + ε...
Ensemble-Methods 2004/02/27 Predictions (2008)
We assume that the distribution of a response variable y depends on a vector of input variables x through a function f: Special cases are: D(y|x) = D(y|f(x)). y = f(x) + ε (Regression) P (y =...
Boosting Additive Models using Component-wise P-Splines (2008)
Matthias Schmid, Torsten Hothorn, Matthias Schmid, Torsten Hothorn
We consider an efficient approximation of Bühlmann & Yu’s L2Boosting algorithm with component-wise smoothing splines. Smoothing spline base-learners are replaced by P-spline base-learners...
Achim Zeileis, Torsten Hothorn, Kurt Hornik
Recursive partitioning is embedded into the general and well-established class of parametric models that can be fitted using M-type estimators (including maximum likelihood). An algorithm for...
x y Zweistichproben-t-Test: P = 0.033 Welch-t-Test: P = 0.034 (2008)
X und Y- aber das experimentelle Design hilft hier weiter!
Peter Bühlmann, Torsten Hothorn, Eth Zürich
We present a statistical perspective on boosting. Special emphasis is given to estimating potentially complex parametric or nonparametric models, including generalized linear and additive models as...
News The Newsletter of the R Project Volume 7/2, October 2007 Editorial (2008)
Welcome to the October 2007 issue of R News, the second issue for this year! Last week, R 2.6.0 was released with many new useful features: dev2bitmap allows semi-transparent colours to be used,...
News The Newsletter of the R Project Volume 7/1, April 2007 Editorial (2008)
follows the release of R version 2.5.0. This major revision, in addition to many other features, brings better support of JAVA and Objective C to our desks. Moreover, there is a new recommended...
News The Newsletter of the R Project Volume 7/3, December 2007 Editorial (2008)
Shortly before the end of 2007 it’s a great pleasure for me to welcome you to the third and Christmas issue of R News. Also, it is the last issue for me as editorial board member and before John...
Flexible Boosting of Accelerated Failure Time Models (2008)
Schmid, Matthias, Hothorn, Torsten
When boosting algorithms are used for building survival models from high-dimensional data, it is common to fit a Cox proportional hazards model or to apply semiparametric least squares techniques....
Simultaneous Inference in General Parametric Models (2008)
Hothorn, Torsten, Bretz, Frank, Westfall, Peter
Simultaneous inference is a common problem in many areas of application. If multiple null hypotheses are tested simultaneously, the probability of rejecting erroneously at least one of them increases...
Implementing a Class of Permutation Tests: The coin Package (2008)
Torsten Hothorn, Kurt Hornik, Achim Zeileis
The R package coin implements a unified approach to permutation tests providing a huge class of independence tests for nominal, ordered, numeric, and censored data as well as multivariate data at...
Simultaneous Inference in General Parametric Models (2008)
Torsten Hothorn, Frank Bretz, Peter Westfall, Torsten Hothorn, Frank Bretz, Novartis Pharma Ag, ...
Simultaneous inference is a common problem in many areas of application. If multiple null hypotheses are tested simultaneously, the probability of rejecting erroneously at least one of them increases...
References any Linare Algebra book Examples X <- cbind(1, diag(10)) # reduced rank mpinv(X) 1 2 multilm multilm Mulivariate Linear Models multilm Description multilm fits a multivariate linear...
Bundling predictors in R (2007)
The construction of a good classifier based on a learning sample can be seen as a three step procedure. At first, we use the observations in the learning sample to construct different rules. In the...
MGE Mean Gene Expression (2007)
Torsten Hothorn, Maintainer Torsten Hothorn
Description Maximally selected rank and Gauss statistics with several p-value
StatDataML--An XML Format for (2007)
Statistical Data David, David Meyer, Friedrich Leisch, Torsten Hothorn, Kurt Hornik, In Cooperation
In order to circumvent common difficulties in exchanging statistical data between heterogeneous applications (format incompatibilities, technocentric data representation), we introduce an XML-based...
Bias in random forest variable importance measures: Illustrations, sources and a solution (2007)
Strobl, Carolin, Boulesteix, Anne-Laure, Zeileis, Achim, Hothorn, Torsten
Abstract Background Variable importance measures for random forests have been receiving increased attention as a means of variable selection in many classification tasks in bioinformatics and related...
Generalized Maximally Selected Statistics (2007)
Hothorn, Torsten, Zeileis, Achim
Maximally selected statistics for the estimation of simple cutpoint models are embedded into a generalized conceptual framework based on conditional inference procedures. This powerful framework...
Generalized Maximally Selected Statistics (2007)
Hothorn, Torsten; Zeileis, Achim
Maximally selected statistics for the estimation of simple cutpoint models are embedded into a generalized conceptual framework based on conditional inference procedures. This powerful framework...
Implementing a Class of Permutation Tests: The coin Package (2007)
Hothorn, Torsten, Hornik, Kurt, Van De Wiel, Mark A., Zeileis, Achim
The R package coin implements a unified approach to permutation tests providing a huge class of independence tests for nominal, ordered, numeric, and censored data as well as multivariate data at...
Let's Have a party! An Open-Source Toolbox for Recursive Partytioning (2007)
Hothorn, Torsten, Zeileis, Achim, Hornik, Kurt
Package party, implemented in the R system for statistical computing, provides basic classes and methods for recursive partitioning along with reference implementations for three recently-suggested...
Boosting Additive Models using Component-wise P-Splines (2007)
Schmid, Matthias, Hothorn, Torsten
We consider an efficient approximation of Bühlmann & Yu’s L2Boosting algorithm with component-wise smoothing splines. Smoothing spline base-learners are replaced by P-spline base-learners which...
Variable Selection and Model Choice in Geoadditive Regression Models (2007)
Kneib, Thomas, Hothorn, Torsten, Tutz, Gerhard
Model choice and variable selection are issues of major concern in practical regression analyses. We propose a boosting procedure that facilitates both tasks in a class of complex geoadditive...
SystemRequirements Java (> = 5.0) (2007)
Kurt Hornik, Torsten Hothorn, David Meyer, Achim Zeileis
Description An R interface to Weka (Version 3.5.7). Weka is a collection of machine learning algorithms for data mining tasks written in Java, containing tools for data pre-processing,...
SystemRequirements Java (> = 5.0) (2007)
Kurt Hornik, Torsten Hothorn, David Meyer, Achim Zeileis
Description An R interface to Weka (Version 3.5.6). Weka is a collection of machine learning algorithms for data mining tasks written in Java, containing tools for data pre-processing,...
1 L2-Boost and Infinitesimal Forward Stagewise Linear Regression (2007)
Peter Bühlmann, Torsten Hothorn, Trevor Hastie
We congratulate the authors (hereafter BH) for an interesting take on the boosting technology, and for developing a modular computational environment in R for exploring their models. Their use of...
Evaluating Model-based Trees in Practice (2006)
Zeileis, Achim, Hothorn, Torsten, Hornik, Kurt
A recently suggested algorithm for recursive partitioning of statistical models (Zeileis, Hothorn and Hornik, 2005), such as models estimated by maximum likelihood or least squares, is evaluated in...
Bias in Random Forest Variable Importance Measures: Illustrations, Sources and a Solution (2006)
Strobl, Carolin, Boulesteix, Anne-Laure, Zeileis, Achim, Hothorn, Torsten
Variable importance measures for random forests have been receiving increased attention as a means of variable selection in many classification tasks in bioinformatics and related scientific fields,...
Permutation Tests for Structural Change (2006)
Zeileis, Achim, Hothorn, Torsten
The supLM test for structural change is embedded into a permutation test framework for a simple location model. The resulting conditional permutation distribution is compared to the usual...
Bias in Random Forest Variable Importance Measures: Illustrations, Sources and a Solution (2006)
Strobl, Carolin, Boulesteix, Anne-Laure, Zeileis, Achim, Hothorn, Torsten
Variable importance measures for random forests have been receiving increased attention as a means of variable selection in many classification tasks in bioinformatics and related scientific fields,...
Spatial Smoothing Techniques for the Assessment of Habitat Suitability (2006)
Kneib, Thomas, Müller, Jörg, Hothorn, Torsten
Precise knowledge about factors influencing the habitat suitability of a certain species forms the basis for the implementation of effective programs to conserve biological diversity. Such knowledge...
Torsten Hothorn, Peter Bühlmann, Rine Dudoit, Annette Molinaro
This document reproduces the data analyses presented in Hothorn et al. (2006). For a description of the theory behind applications shown here we refer to the original manuscript. 1.1 Acute myeloid...
Version 0.2-4 Date 2006-05-07 Title R/Weka interface The RWeka Package (2006)
Kurt Hornik, Torsten Hothorn, David Meyer, Achim Zeileis
Description An R interface to Weka (Version 3.5.2). Weka is a collection of machine learning algorithms for data mining tasks written in Java, containing tools for data pre-processing,...
Model-based boosting in high dimensions (2006)
Hothorn, Torsten, Bühlmann, Peter
Summary: The R add-on package mboost implements functional gradient descent algorithms (boosting) for optimizing general loss functions utilizing componentwise least squares, either of parametric...
Hothorn, Torsten, Bühlmann, Peter, Dudoit, Sandrine, Molinaro, Annette, Van Der Laan, Mark J.
We propose a unified and flexible framework for ensemble learning in the presence of censoring. For right-censored data, we introduce a random forest algorithm and a generic gradient boosting...
For information about other BioMed Central publications go to (2006)
Ludwig A Hothorn, Michael Vaeth, Torsten Hothorn, Michael Vaeth B, Torsten Hothorn C
This Provisional PDF corresponds to the article as it appeared upon acceptance. Fully formatted PDF and full text (HTML) versions will be made available soon. Trend tests for the evaluation of...
Hothorn, Torsten, Buhlmann, Peter, Dudoit, Sandrine, Molinaro, Annette M., Van Der Laan, Mark J.
We propose a unified and flexible framework for ensemble learning in the presence of censoring. For right-censored data, we introduce a random forest algorithm and a generic gradient boosting...
Hothorn, Torsten, Buhlmann, Peter, Dudoit, Sandrine, Molinaro, Annette M., Van Der Laan, Mark J.
We propose a unified and flexible framework for ensemble learning in the presence of censoring. For right-censored data, we introduce a random forest algorithm and a generic gradient boosting...
Hothorn, Torsten, Buhlmann, Peter, Dudoit, Sandrine, Molinaro, Annette M., Van Der Laan, Mark J.
We propose a unified and flexible framework for ensemble learning in the presence of censoring. For right-censored data, we introduce a random forest algorithm and a generic gradient boosting...
Hothorn, Torsten, Buhlmann, Peter, Dudoit, Sandrine, Molinaro, Annette M., Van Der Laan, Mark J.
We propose a unified and flexible framework for ensemble learning in the presence of censoring. For right-censored data, we introduce a random forest algorithm and a generic gradient boosting...
Model-based recursive partitioning (2005)
Zeileis, Achim, Hothorn, Torsten, Hornik, Kurt
Recursive partitioning is embedded into the general and well-established class of parametric models that can be fitted using M-type estimators (including maximum likelihood). An algorithm for...
A Lego System for Conditional Inference (2005)
Hothorn, Torsten, Hornik, Kurt, Zeileis, Achim
Conditioning on the observed data is an important and flexible design principle for statistical test procedures. Although generally applicable, permutation tests currently in use are limited to the...
The design and analysis of benchmark experiments (2005)
Torsten Hothorn, Friedrich Leisch, Kurt Hornik, Achim Zeileis, ...
The assessment of the performance of learners by means of benchmark experiments is an established exercise. In practice, benchmark studies are a tool to compare the performance of several competing...
Unbiased recursive partitioning: A conditional inference framework (2005)
Torsten Hothorn, Kurt Hornik, Achim Zeileis
Recursive binary partitioning is a popular tool for regression analysis. Two fundamental problems of exhaustive search procedures usually applied to fit such models have been known for a long time:...
Hothorn, Torsten, Bühlmann, Peter, Dudoit, Sandrine, Molinaro, Annette, Van Der Laan, Mark J.
We propose a unified and flexible framework for ensemble learning in the presence of censoring. For right-censored data, we introduce a random forest algorithm and a generic gradient boosting...
Bioconductor: open software development for computational biology and bioinformatics (2004)
Gentleman, Robert C, Carey, Vincent J, Bates, Douglas M, Bolstad, Ben, Dettling, Marcel, Dudoit, Sandrine, ...
Abstract The Bioconductor project is an initiative for the collaborative creation of extensible software for computational biology and bioinformatics. The goals of the project include: fostering...
Bioconductor: Open software development for computational biology and bioinformatics (2004)
Gentleman, Robert C., Carey, Vincent J., Bates, Douglas J., Bolstad, Benjamin M., Dettling, Marcel, Dudoit, Sandrine, ...
The Bioconductor project is an initiative for the collaborative creation of extensible software for computational biology and bioinformatics. We detail some of the design decisions, software...
Bioconductor: Open software development for computational biology and bioinformatics (2004)
Gentleman, Robert C., Carey, Vincent J., Bates, Douglas J., Bolstad, Benjamin M., Dettling, Marcel, Dudoit, Sandrine, ...
The Bioconductor project is an initiative for the collaborative creation of extensible software for computational biology and bioinformatics. We detail some of the design decisions, software...
Bioconductor: Open software development for computational biology and bioinformatics (2004)
Gentleman, Robert C., Carey, Vincent J., Bates, Douglas J., Bolstad, Benjamin M., Dettling, Marcel, Dudoit, Sandrine, ...
The Bioconductor project is an initiative for the collaborative creation of extensible software for computational biology and bioinformatics. We detail some of the design decisions, software...
Bioconductor: Open software development for computational biology and bioinformatics (2004)
Gentleman, Robert C., Carey, Vincent J., Bates, Douglas J., Bolstad, Benjamin M., Dettling, Marcel, Dudoit, Sandrine, ...
The Bioconductor project is an initiative for the collaborative creation of extensible software for computational biology and bioinformatics. We detail some of the design decisions, software...
Unbiased Recursive Partitioning: A Conditional Inference Framework (2004)
Hothorn, Torsten, Hornik, Kurt, Zeileis, Achim
Recursive binary partitioning is a popular tool for regression analysis. Two fundamental problems of exhaustive search procedures usually applied to fit such models have been known for a long time:...
Bundling classifiers with an application to glaucoma diagnosis (2003)
The combination of classifiers of arbitrary type, for example classification trees,linear discriminant analysis, nearest neighbors or the logistic regressionmodel, is most desirable for at least two...
The design and analysis of benchmark experiments (2003)
Hothorn, Torsten, Leisch, Friedrich, Zeileis, Achim, Hornik, Kurt
The assessment of the performance of learners by means of benchmark experiments is established exercise. In practice, benchmark studies are a tool to compare the performance of several competing...
Bundling classifiers with an application to glaucoma diagnosis / (2003)
Dortmund, University, Diss., 2003 (Nicht für den Austausch).
Economics and Management Science’). The Design and Analysis of Benchmark Experiments (2003)
Torsten Hothorn, Friedrich Leisch, Achim Zeileis, Kurt Hornik, In Cooperation, Torsten Hothorn, ...
Papers published in this report series are preliminary versions of journal articles. This piece of research was supported by the Austrian Science Foundation
Meyer, David, Leisch, Friedrich, Hothorn, Torsten, Hornik, Kurt
In order to circumvent common difficulties in exchanging statistical data between heterogeneous applications (format incompatibilities, technocentric data representation), we introduce an XML-based...
Double-bagging: combining classi ers by bootstrap (2002)
Torsten Hothorn, Berthold Lausen
www.elsevier.com/locate/patcog
Bioconductor: open software development for computational biology and bioinformatics
Gentleman, Robert C, Carey, Vincent J, Bates, Douglas M, Bolstad, Ben, Dettling, Marcel, Dudoit, Sandrine, ...
A detailed description of the aims and methods of the Bioconductor project, an initiative for the collaborative creation of extensible software for computational biology and bioinformatics.
Bioconductor: open software development for computational biology and bioinformatics
Gentleman, Robert C, Carey, Vincent J, Bates, Douglas M, Bolstad, Ben, Dettling, Marcel, Dudoit, Sandrine, ...
A detailed description of the aims and methods of the Bioconductor project, an initiative for the collaborative creation of extensible software for computational biology and bioinformatics.
Bias in random forest variable importance measures: Illustrations, sources and a solution
Strobl, Carolin, Boulesteix, Anne-Laure, Zeileis, Achim, Hothorn, Torsten
A Lego System for Conditional Inference
Hothorn, Torsten, Hornik, Kurt, Van De Wiel, Mark A., Zeileis, Achim
Boosting additive models using component-wise P-Splines
Schmid, Matthias, Hothorn, Torsten
An efficient approximation of L2 Boosting with component-wise smoothing splines is considered. Smoothing spline base-learners are replaced by P-spline base-learners, which yield similar prediction...
Hothorn, Ludwig A, Vaeth, Michael, Hothorn, Torsten
One possibility for the statistical evaluation of trends in epidemiological exposure studies is the use of a trend test for data organized in a 2 × k contingency table. Commonly, the exposure data...