Home

Tuesday, 3 February, 2026

Master´s Thesis Defense Presentations

Room #320

12:00    

Sosa Galindo Ivan Israel: Regional Growth, Internal Migration, and Persistent Inequality in Mexico

Chair: Byeongju Jeong

Opponent: Stephanie Ettmeier

13:00    

Lark Madison: Bitcoin Market Activity and Public Attention to Central Bank Digital Currencies (2017-2024)

Chair: Branislav Saxa

Opponent: Jan Zápal

14:00    

Ďurian Ladislav: Estimating Genre-Specific Demand Elasticities in the Amazon Book Market: A Double/Debiased Machine Learning Approach

Chair: Paolo Zacchia

Opponent: Klaus Achim Ahrens

Ladislav Ďurian

Abstract:

In this thesis, I examine how demand elasticities with respect to price, star ratings, and newspaper reviews vary across book genres in the Amazon book marketplace. Using daily panel data from Amazon domains in the United States, the United Kingdom, and Canada in 2018, I estimate a benchmark linear fixedeffects model and contrast it with a double/debiased machine learning specification that flexibly accounts for the nonlinear effects of control variables. I find substantial heterogeneity in consumer demand across genres and suggest how marketers can exploit this heterogeneity when designing genre-based segmentation strategies. Beyond its marketing implications, the thesis offers a methodological insight by showing that double/debiased machine learning yields genre-specific elasticity estimates that are robust to nonlinear confounding, unlike linear fixed-effects models.

Full Text: “Estimating Genre-Specific Demand Elasticities in the Amazon Book Market: A Double/Debiased Machine Learning Approach"