Staff and
Department Search

Sebastian Büscher

Photo of the person Herr Sebastian Büscher - Opens the image in full size on a new page

Contact

1. Faculty of Business Administration and Economics / Lehrstuhl für Ökonometrie

Wiss. Mitarbeiter bei Prof. Dr. Bauer

E-Mail
sebastian.buescher@wiwi.uni-bielefeld.de  
Phone
+49 521 106-4895  
Fax
+49 521 106- 6425 
Phone No. of Secretary
+49 521 106-6946 Secretary's Details
Office
UHG V9-108 Locations Map
Postbox
1734

Curriculum Vitae

Academic Education

since 2018 Doctoral Studies in Economic Sciences, Econometrics

2015-2017 Erasmus Mundus Joint Master Degree Programme MathMods - Mathematical Modelling in Engineering at the Universities

  • Università degli Studi dell'Aquila, Italy
  • University of Hamburg, Germany
  • Universitat Autònoma de Barcelona, Spain
  • Thesis: Development of a mathematical model to describe the liquid exchange in a washing machine during the washing and rinsing process in cooperation with Miele & Cie. KG

2011-2015 Bachelor of Science in Mathematics and Physics at Bielefeld University

  • 2013-2014 at the University of Edinburgh with Erasmus Programme
  • Thesis: Der Satz von Riesz-Schauder



Publication

  • Büscher, S; Bauer, D (2024): Weighting strategies for pairwise composite marginal likelihood estimation in case of unbalanced panels and unaccounted autoregressive structure of the errors, Transportation Research Part B: Methodological, Volume 181, https://doi.org/10.1016/j.trb.2024.102890.
  • Büscher, S; Batram, M; Bauer, D (2019): Using Motifs for Population Synthesis in Multi-Agent Mobility Simulation Models, Stochastic Models, Statistics and Their Applications, 335-349.

Talks

  • Büscher, S; Bauer, D (2024): Lagrange multiplier type test to detect structural intra-personal heterogeneity in Composite Marginal Likelihood estimation of discrete choice models, talk at the 8th International Choice Modelling Conference, Puerto Varas, Chile
  • Büscher, S (2023): How to test for intra-personal heterogeneity and temporal effects in discrete choice models in an economic manner, talk at the 12th Young Researchers Workshop of the Centre for Statistics, Bielefeld
  • Büscher, S (2022): Robust estimation of misspecified discrete choice models, talk at the 11th Young Researchers Workshop of the Centre for Statistics, Bielefeld
  • Büscher, S; Bauer, D (2022): Weighting strategies for pairwise composite marginal likelihood estimation in case of unbalanced panels and unaccounted autocorrelation of the errors, talk at the 7th International Choice Modelling Conference, Reykjavik, Iceland
  • Büscher, S (2022): Weighting strategies for pairwise composite marginal likelihood estimation, talk at the 10th Young Researchers Workshop of the Centre for Statistics, Bielefeld
  • Büscher, S (2021): Econometrics vs. Machine Learning: Who is better at predicting human mobility patterns?, talk at the 9th Young Researchers Workshop of the Centre for Statistics, Bielefeld
  • Büscher, S (2019): Modeling Motif Choice: Using Out of Sample Prediction Performance to Compare Different Discrete Choice Models for Model Selection to Analyze Temporal Stability, talk at the 8th Young Researchers Workshop of the Centre for Statistics, Bielefeld
  • Büscher, S; Bauer, D; Batram, M (2019): Modeling Motif Choice: Comparing the prediction performance of different discrete choice models to analyze temporal stability, talk at the 6th International Choice Modelling Conference, Kobe, Japan
  • Büscher, S (2019): Using Motifs in person-centred mobility simulation models, talk at the 7th Young Researchers Workshop of the Centre for Statistics, Bielefeld
  • Büscher, S; Batram, M; Bauer, D (2019): Using Motifs for Population Synthesis in Multi-Agent Mobility Simulation Models, talk at the 14th Workshop on Stochastic Models, Statistics and their Application, Dresden
  • Büscher, S (2018): An analysis of two decades of German mobility data with an emphasis on temporal stability using Multinomial Logit Models, talk at the 6th Young Researchers Workshop of the Centre for Statistics, Bielefeld
  • Büscher, S (2018): The MaCML approach to probit model estimation with application on motif choice, talk at the 5th Young Researchers Workshop of the Centre for Statistics, Bielefeld

Current research topics

Discrete Choice Modells
In my research, I mainly deal with discrete choice models. This involves modelling the choice probabilities of a finite set of choices. Classical applications include transport mode choice, political choice, or product choice.

In particular, I deal with the composite marginal likelihood (CML) estimation of so-called probit models. In CML estimation, the product of the probabilities of pairs of observations is used instead of the joint probability of all observations for a person. Doing so reduces the necessary computational effort considerably. This comes at the expense of the statistical efficiency of the estimator.

My current focus is on
1. reducing the loss of efficiency by using weighting strategies,
2. making the estimator more robust to misspecification through the usage of weighting strategies, and
3. testing for time dependencies within the model using the gradient contributions of the observation pairs.