Direkt zum InhaltDirekt zur SucheDirekt zur Navigation
▼ Zielgruppen ▼

Humboldt-Universität zu Berlin - Emmy Noether Research Group

Topics for Theses


 

  • Deep distributional learning
  • Standard errors of Variational Inference
  • Uncertainty quantification for deep learning
  • Bayesian stacking approaches
  • Bayesian neural network structures
  • Deep Gaussian process modelling
  • Multiple Imputation for Panel Data
  • Using Stacking to Average Distributional Regression Models
  • Bivariate Distributional Regression for Wind Speed and Wind Direction
  • Modelling Income Inequality with Semiparametric Transformation Models
  • Bayesian Nonparametric Conditional Density Estimators
  • Distributional Joint Modelling
  • Probabilistic Weather Forecasts
  • Approximations of Normalizing Constants in Doubly-Intractable Likelihoods
  • Effect Fusion of Categorial Predictors
  • Effect Selection in Semiparametric Quantile Regression Models
  • Measuring the Explained Variance in Structured Additive Distributional Regression
  • Bayesian Hierarchical Modelling of Hedonic Housing Prices
  • Comparisons and Implementation of Non-Local Shrinkage Priors
  • Predictive information criteria from a Bayesian perspective

 

...get in touch with us: stat@wiwi.hu-berlin.de

 

 

Completed Theses


 

Master's Theses

  • Kang Yang (2020): Deep Generalised Additive Regression Models for Location, Scale and Shape
  • Paulina Kurowska (2020): Bayesian Variable Selection Approach in Additive Quantile Regression
  • Patricia Craja (2020): Fraud Detection in financial Statements using Deep Learning based Natural Language Processing
  • Andrii Zakharov (2019): Development of a Customer Insolvency Prediction Model following the Cross-Industry Standard Process for Data Mining (CRISP-DM)
  • Jan Reher (2019): Purchase Prediction with Topic Models Using Variational Inference
  • Lukas Kemkes (2019): Comparison and combination of extractive and abstractive techniques for machine text summarization
  • Yinan Wu (2019): Bayesian Adaptive Lasso for Zero Inflated Count Model and Overdispersion
  • Jerome Bau (2019): Context-aware Sentence Generation from Keywords
  • Stanislaus Stadlmann (2017): bamlss.vis: An R Package to Interactively Analyze and Visualize Bayesian Additive Models for Location, Scale and Shape (BAMLSS) Using the Shiny Framework
  • Hannes Riebl (2017): Multivariate Quantile Residuals in Copula Regression: Model Evaluation and Selection
  • Manuel Carlan (2016): Bayesian Variable Selection in Semiparametric Regression
  • Paul Wiemann (2015): Statistische Analyse von Advice-Taking
  • Manuel Vonrueti (2015): Hyperparameterwahl einer neuartigen Spike und Slab Priori für Verteilungsregression
  • Anastasia Gorbunova (2015): Statistische Analyse von länderspezifischen Energiemixen

Bachelor's Theses

 
  • Elisabeth Krawczyk (2019): Comparing applicability of prevalent Clustering Algorithms for Document Clustering
  • Hannes Riebl (2014): Power Law Verteilungen in Generalisierten Additiven Modellen für Lokation, Skala und Form
  • Philipp Hanheide (2013): Messung der Armuts-Vulnerabilität