Topics for Theses in Statistics

The following list comprises subjects that are currently available for theses in statistics. Subjects suitable for bachelor theses are marked as BA while topics that are more suitable for master theses are marked as MA. Of course you can also suggest your own topic for a thesis if the topic comprises a strong focus on quantitative methods.


  • Defining and analyzing essential work (BA/MA)
  • Nutzenoptimierung mehrerer Güter unter Nebenbedingungen durch Anwendung der Methode der Lagrange-Multiplikatoren (BA)
  • Educational Decisions and Social Inequality (BA/MA)
  • Bayesian spatial models with Gaussian Markov Random fields (MA)
  • Bayesian spatial model for circular data with independent projections of the angular quantity (MA)
  • Evaluating models for German annual accumulated precipitation (BA, english)
  • Analyzing crime data with nearest neighbours approaches (BA, english)
  • Analyse von kommunalen/betrieblichen Umwelt- und Energiemanagementsystemen (BA)
  • Pro-Environmental Workplace Intention Behavior at the University (BA/MA)
  • Implementation of a diagnostic tool for misspecification of random effects in mixed models based on the gradient function (MA)
  • Simulation assessment of the normal approximation for (multiple) estimators in expectile regression (BA/MA)
  • Two-stage imputation of missing values (BA)
  • Jackknife and the common mean problem (BA/MA)
  • Data Literacy: Data Wrangling und plotten mit R und Python. Ein Vergleich (BA)
  • Deep Learning based additive models (MA)
  • Regression und Eugenik: Waren Galton, Pearson und Fisher Rassisten? (BA)
  • Bayesian Restricted Spatial Regression for Dealing with Spatial Confounding (MA)
  • Applications of techniques for anomaly detection on data loading processes (MA, in collaboration with CEWE Stiftung & Co. KGaA, Oldenburg)
  • Random Number Generation from the Convolution of Random Variables
  • Endogeneity in Stochastic Frontier Analysis
  • Variable selection for causal inference in observational data: state-of-the-art practices and propositions for GAMLSS (BA/MA)
  • GAMLSS in regression discontinuity designs: methods and applications in impact evaluation (MA)
  • Analyses of pedestrian frequency data considering explanatory covariates like weather (BA)
  • Employing Variational Approximation in Structured Additive Regression (MA)
  • The functional horseshoe prior (MA)
  • Exploring the CaptitalBikeData with spatio-temporal models (MA)
  • R packages surrounding distributional regression (MA)
  • Evaluating parametric and nonparametric transformation models (BA/MA)
  • Analyzing income (inequalities) with conditional transformation models (BA/MA)
  • Natural language processing: dynamic and structural topic models (MA)
  • Causal Inference and Machine Leanring: Estimation of heterogeneous treatment effects using random forests (MA)
  • Causal Inference and Deep Learning: Deep instrumental variables networks (MA)
  • Spatial modeling of geocoded Latent Dirichlet Allocation based topic distributions with Gaussian Markov Random Fields(MA)
  • Supervised classification of documents with machine learning and deep learning (MA)
  • Word embeddings in topic models (e.g. Word2Vec and LDA) (MA)
  • Prediction of terror attacks with machine learning and deep learning models (e.g. XGBoost and LSTM) (MA)