Springe direkt zu Inhalt

Advanced Methods for Analyzing Discrete Survey Answers


This course focuses on the estimation of models with discrete dependent variables, presenting a rigorous academic exploration of statistical methods and their application to real-world datasets. Through an examination of maximum likelihood estimation and various models, including linear regression, binary models (logit and probit), multinomial models, and count data analysis, students will gain a comprehensive understanding of the complexities inherent in analyzing discrete outcomes. Hands-on exercises using R software will provide practical experience in model implementation and interpretation, enabling students to conduct sophisticated data analysis with precision and confidence.

Table of contents:

  • Introduction
  • Maximum Likelihood
  • Linear Models
  • Binary Models
  • Multinomial Models
  • Models for Count Data
  • Generalized Models for Discrete Outcomes



  • Duration: 120 minutes
  • Components
    • Estimation of three discussed models in R with provided data sets (75% of the final grade)
    • Interpretation and theoretical questions (25% of the final grade)

The examination will assess your proficiency in applying the discussed models using R, which constitutes the majority (75%) of the assessment. Additionally, questions will be designed to evaluate your ability to interpret results and engage with theoretical concepts, contributing to the remaining 25% of the overall grade.

Forschungsschwerpunkt Statistik und Ökonometrie