Humboldt-Universität zu Berlin - Statistics

Chair of Statistics - Final theses

If you are interested in writing your thesis at the Chair of Statistics, please contact us for more details.


Useful links:


Student theses:

Author Title Supervisor


Qingzi Huo

An Extension of the Generalized Covariance Measure for Conditional Independence Testing

Sonja Greven




Chenyin Zhao

Comparison of Covariance Estimation Methods for Sparse Functional Data

Sonja Greven

Johannes Brandau

Invariant Feature Learning via Orthogonalization

Sonja Greven

Vincent Karpf

The Necessity of Complex Data: Object Importance by Removing for High-Dimensional Structured Objects

Sonja Greven

David Frohnau

Using Diffusion Models to Create Training Data for Image Classification Algorithms

Sonja Greven
Ana Karina Enriquez de los Rios

Functional Spatial Correlation Analysis on Crime Data of Mexico City

Sonja Greven

Maximilian Hildebrand

Blood Glucose Forecasting

Nadja Klein

Stanislav Slowinski

Der Einfluss der europäischen Struktur- und Investitionsfonds im Kontext des Problems der Heterogenität der Entwicklung in der Europäischen Union

Sonja Greven

Till Bethge

Bayesian Variable Selection for Ordinal Regression with an Application to Gene Expression Data

Nadja Klein

Johannes Feeser

On Estimating Causal Effects in Poisson-based Density Regression with an Application to the Retirement-Consumption Puzzle

Sonja Greven
Johannes Abele

Modelling Disease Severity using Semi- and Weakly Supervised Variational Autoencoders

Nadja Klein
Wei Ji Tan

Using the Horseshoe Prior for the Estimation of Non-linear Functions

Nadja Klein
Benedikt Lütke-Schwienhorst

Dropout Regularization in Generalized Linear Models Based on Double Exponential Families

Nadja Klein


Till Baldenius

Analysing Heat and Experienced Racial Segregation Using Large-Scale Foot Traffic Data

Nadja Klein

Ákos Blaskovics

Application of Regression Trees on Compositional Data Using European Parliament Election Results

Sonja Greven
Julian Schneider Extension of Treatment Effect Analysis to Dynamic Networks: An Application to PrEP in the context of HIV Transmission Sonja Greven
Friedemann Brockmeyer  Summary Characteristics for Real-valued Marked Point Patterns on Linear Networks Sonja Greven

Marco Simnacher

Extension of Generalized Additive Models to Graph-valued Data with Application to Covid-19 Impacts on European Air Transportation

Sonja Greven

Katja Weimer

Inference for Density-on-Scalar Regression in Bayes Hilbert Spaces

Sonja Greven
Henry Böddeker

Sizing Flags: Multi-Class Size Recommnedations for Fashion E-Commerce

Nadja Klein
Lucrece Mefowe Fokam

Correcting Underreporting in German Covid-19 Data

Sonja Greven
Felix Germaine

Interpretable Modelling of ICU Patients Remaining Length-of-Stay Distribution using Tabular Patient Data, Clinical Notes and Irregularly Spaced Clinical Measurement

Nadja Klein


Bianca Neubert

Application of Hybrid Multivariate Functional Principal Component Analysis for the Analysis of Multivariate Spatial Point Process Summary Characteristics

Sonja Greven
Julius Freidank

Vergleich von Vorhersagemodellen zu Stornierungen von Hotelbuchungen

Sonja Greven
Jana Kleinemeier Using Variational Inference to Estimate Structred Additive Distributional Regression Models Nadja Klein
Anna Semenova Reconstructing Multivariate Functional Data with Medical Applications Sonja Greven
Manuel Pfeuffer Elastic Full Procrustes Means for Sparse and Irregular Curves Sonja Greven
Gesa Julia Kröger

Scalar-on-Composition Regression to evaluate the Impact of Class Competition on Educational achievement

Sonja Greven
Rodrigo Vazquez Anomaly Data Quality Detection using Deep Autoencoders Nadja Klein
Jost von Petersdorff-Campen Copula Regression for Discrete Data – An Application to Covid-19 Nadja Klein
Yuliya Vandysheva Die Modellierung der COVID-19-Fallzahlen in Abhängigkeit von Strukturdaten zu Wetter und Bevölkerung in Berlin Nadja Klein
Chris Kolb Application of Deep Distributional Regression to the Geospatial Distribution of Neglected Tropical Disease Vectors Nadja Klein
Per Joachims Uncertainty Quantification with Bayesian Neural Networks Nadja Klein
Malgorzata Olesiewicz Modelling of Missing not at Random Corporate Emissions Data for Sustainable Investment Sonja Greven
Clara Hoffmann Marginally Calibrated Posterior Densities for End-to-End Learning in Autonomous Driving Nadja Klein
Michael Schimpke Sentiment-Optimised Translation: An Approach for Multilingual Sealing of Pre-Trained Sentiment Classifiers Nadja Klein
Jakub Kondek Classification of Empathic and Non-Empathic Emails using Prototypical Networks Nadja Klein
Silvia Ventoruzzo On the Role of Weather in Predicting Online Sales of Different Product Categories Nadja Klein
Safiul Alom Uncertainty in Machine-Learning-based E-Mail Classification for Improved Customer Service Nadja Klein
Lukas Mödl The RODEO Approach for Nonparametric Density Estimation Nadja Klein
Anna Elisabeth Riha Hyperpriorsensitivity of Bayesian Wrapped Gaussian Processes with an Application to Wind Data Nadja Klein
Aaron Burgemeister Benchmarking Different Machine Learning Methods Against Logistic Regression in Credit Scoring Sonja Greven
Xiaoji Du Analysis of a Restaurant Database of 31 European Cities Sonja Greven
Jil Kollmus-Heege

Dynamic Prediction in Flexible Bayesian Additive Joint Models

Sonja Greven
Sisi Huang An Image Classification Tool of Wikimedia Commons Sonja Greven
Thomas Siskos Career Recommendations using Supervised Latent Dirichlet Allocation Sonja Greven
Kang Yang Deep Generalised Additive Regression Models for Location, Scale and Shape Nadja Klein
Paulina Kurowska Bayesian Variable Selection Approach in Additive Quantile Regression Nadja Klein
Yinan Wu Bayesian Adaptive Lasso for Zero Inflated Count Model and Overdispersion Nadja Klein
Jerome Bau Context-aware Sentence Generation from Keywords Nadja Klein
Lukas Kemkes Comparison and Combination of Extractive and Abstractive Techniques for Machine Text Summarization Nadja Klein
Andrii Zakharov Development of a Customer Insolvency Prediction Model following the Cross-Industry Standard Process for Data Mining (CRISP-DM) Nadja Klein


Selected Theses:

Bachelor Theses

  • Regressionsmodelle und Hauptkomponentenanalyse für Billard-Bewegungstrajektorien (2019)
  • Classification of input devices using cursor trajectories (2017)
  • Funktions-auf-Skalar-Regression für Zellchipdaten (2015)
  • The Correlation Coefficient as a Measure of Similarity for Thermal Images in Medicine (2014)
  • Vergleich von Selektionskriterien für die Anzahl funktionaler Hauptkomponenten (2011)


Master Theses

  • A comparison of different approaches to the joint modelling of longitudinal and time-to-event outcomes (2019)
  • Multivariate functional regression models (2018)
  • Boosting flexible joint models for longitudinal and time-to-event data with medical applications
  • Clustering räumlicher Kurven zur Identifikation von Laufmustern bei Hypermarathonläufern (2017)
  • Joint modelling of longitudinal biomarker measurements and time-to-event with an application in
    patients with congestive heart failure (2017)
  • Multivariate Functional Principal Component Regression for Image Data with application to
    Neuroimaging (2017)
  • Elastic Registration and Functional Principal Component Analysis of Curves in 1D and 2D (2017)
  • Boosting Generalized Additive Models for Location, Scale and Shape (GAMLSS) for Functional Data
  • Joint modelling of multiple longitudinal islet autoantibody profiles and the onset of type 1 diabetes
    (2015; Kooperation mit dem Helmholtz-Zentrum München)
  • Klassifikation von Spektren (2016)
  • Modellierung abhängiger/inhomogener Residuen in Regressionsmodellen für funktionale Responses
  • A Comparison of Different Akaike Information Criteria in Linear Mixed Models (2014)
  • Elucidating multivariate effects of various ECG parameters on mortality and evaluation of potentially
    underlying genetic components through genomewide association analyses (2014; Kooperation mit dem
    Helmholtz-Zentrum München)
  • Novelty Detection Methods in High-Dimensional Spaces With An Application On Spectral Data (2014;
    Kooperation mit Siemens)
  • Identifiability in Scalar-on-Functions Regression (2013)
  • Spatio-temporale Aspekte des Screening-Programms für Kolonkarzinome in Bayern (2011; mit Torsten


Diploma Theses

  • Statistischer Performance-Vergleich für Klima-Chemie-Modelle (2012; mit Torsten Hothorn)
  • Conditional Akaike Information Criteria in Generalized Linear Mixed Models (2011; mit Thomas Kneib)