Seminar für angewandte Stochastik   

Im Wintersemester 2016 wird die Professur vertreten durch Christiane Fuchs.

Anschrift / Address

Institut für Statistik
Seminar für angewandte Stochastik
Akademiestraße 1
D-80799 München

Sekretariat (Pia Oberschmidt)

Tel: (+49 89) 2180 2814
Fax: (+49 89) 2180 5308
Prof. Dr. Gerhard Tutz

Newly added:

MÖST, S., PÖßNECKER, W., TUTZ, G. (2015): Variable Selection for Discrete Competing Risks Models. Quality & Quantity, to appear.

TUTZ, G., RAMZAN, S. (2015): Improved Methods for the Imputation of Missing Data by Nearest Neighbor Methods. Computational Statistics & Data Analysis, to appear.

FUCHS, K., GERTHEISS, J., TUTZ, G., (2015): Nearest Neighbor Ensembles for Functional Data with Interpretable Feature Selection. Chemometrics and Intelligent Laboratory Systems, to appear.

GROLL, A., SCHAUBERGER, G., TUTZ, G. (2015): Prediction of major international soccer tournaments based on team-specific regularized Poisson regression: An application to the FIFA World Cup 2014. Journal of Quantitative Analysis of Sports, to appear.

OELKER, M., TUTZ, G. (2015): A General Family of Penalties for Combining Differing Types of Penalties in Generalized Structured Models. Advances in Data Analysis and Classification, published online.

CASALICCHIO, G., TUTZ, G., SCHAUBERGER, G. (2015): Subject-specific Bradley-Terry-Luce Models with Implicit Variable Selection. Statistical Modelling, to appear.


Regression for Categorical
This book introduces basic and advanced concepts of categorical regression with a focus on the structuring constituents of regression. As reference for statisticians, applied researchers, and students it includes many topics not normally included in books on categorical data analysis.

In addition to standard methods such as logit and probit models and their extensions to multivariate settings, the book presents more recent developments in regularized regression with a focus on the selection of predictors; tools for flexible nonparametric regression that yield fits that are closer to the data; non-standard tree-based ensemble methods; and tools for the handling of both nominal and ordered categorical predictors. Issues of prediction are explicitly considered in a chapter that introduces standard and newer classification techniques.