Dr. Moritz Berger

Forschung

Forschungsschwerpunkte

• Regularisierung und Variablenselektion
• Kategoriale Prädiktoren
• Trees
• Item-Response Modelle

Publikationen

• Tutz, Gerhard & Berger, Moritz (2016): Seperating Location and Dispersion in Ordinal Regression Models, Econometrics and Statistics, published online.
  Vorversion: Technical Report 190
• Berger, Moritz & Tutz, Gerhard (2016): Detection of Uniform and Non-Uniform Differential Item Functioning by Item Focussed Trees, Journal of Educational and Behavioral Statistics 41(6), 559–592.
  Vorversion: arXiv:1511.07178
• Tutz, Gerhard & Berger, Moritz (2016): Response Styles in Rating Scales - Simultaneous Modelling of Content-Related Effects and the Tendency to Middle or Extreme Categories, Journal of Educational and Behavioral Statistics 41(3), 239-268.
• Tutz, Gerhard & Berger, Moritz (2015): Item focussed Trees for the Identification of Items in Differential Item Functioning, Psychometrika, 81(3), 727-750.
  Vorversion: arXiv:1502.04329

Dissertation

Berger, Moritz (2016): On the detection of latent structures in categorical data. LMU München: Fakultät für Mathematik, Informatik und Statistik

Preprints

• Bollmann, Stella, Berger, Moritz & Tutz, Gerhard (2016): Item-Focussed Trees for the Detection of Differential Item Functioning in Partial Credit Models, Cornell University Library, arXiv:1609.08970
• Tutz, Gerhard, Schauberger, Gunther & Berger, Moritz (2016): Response Styles in the Partial Credit Model, LMU München, Institut für Statistik, Technical Report 196
• Berger, Moritz & Tutz, Gerhard (2015): Tree-Structured Clustering in Fixed Effects Models, Cornell University Library, arXiv:1512.05169
• Tutz, Gerhard & Berger, Moritz (2014): Tree-Structured Modelling of Categorical Predictors in Regression, Cornell University Library, arXiv:1504.04700

Proceedings

• Berger, Moritz & Tutz, Gerhard (2016): Modelling of Varying Disperion in Cumulative Regression Models, Proceedings of the 31th International Workshop on Statistical Modelling, Volume 1, Rennes.
• Berger, Moritz & Tutz, Gerhard (2015): An Extended Adjacent Categories Model Accounting for Response Styles, Proceedings of the 30th International Workshop on Statistical Modelling, Volume 1, Linz.
  (Best Student Paper)

Vorträge

• A Hierarchy of Cumulative Regression Models to Account for Varying Dispersion, IMBIE, Bonn
• Modelling of Varying Dispersion in Cumulative Regression Models, IWSM 2016, Rennes
• An Extended Adjacent Categories Model Accounting for Response Styles, IWSM 2015, Linz
• Differential Item Functioning with Item focused Trees, Psychoco 2015, Amsterdam
• Regularization of categorical predictors based on trees, DStatG-Nachwuchsworkshop 2014, Hannover

Abschlussarbeiten

• Boosting-Techniken zur Modellierung itemmodifizierender Effekte in Item-Response-Modellen, Master, August 2013 (BestMasters Springer)
• Vergleich von Selektionskriterien für die Anzahl funktionaler Hauptkomponenten, Bachelor, August 2011

R Pakete

• ordDisp (1.0.1): A package to estimate location-shift models or rating-scale models accounting for response styles (RSRS) for the regression analysis of ordinal responses.
• DIFtree (2.1.4): A package to estimate item focussed trees, a method to detect Differential Item Functioning (DIF) based on the Rasch Model or the Logistic Regression Approach for DIF detection.
• structree (1.1.3): A package to perform tree-structured clustering, a method for the fusion of categories of ordinal or nominal predictors or observation units.