CEU Electronic Theses and Dissertations, 2025
| Author | Machavariani, Giorgi |
|---|---|
| Title | Dynamic Decision-Making and Inference Distortions in Behavioral Economics: A Critical Analysis of Over- and Underreaction |
| Summary | This thesis studies systematic deviations from Bayesian updating, focusing on the tendency to overreact to weak signals and underreact to strong ones. I build on the model by Augenblick, Lazarus, and Thaler (Augenblick et al., 2025), presenting a simplified version where individuals form noisy estimates of signal strength and anchor them toward a fixed default. This generates predictable distortions: people overestimate weak signals and underestimate strong ones. The model is theoretically straightforward and shows that asymmetric updating emerges naturally from averaging noisy estimates with an internal anchor. Empirically, the model is validated in three domains: lab experiments, a basketball prediction task, and real-world betting and financial markets. Across all settings, belief updates exceed what is warranted when signals are weak and fall short when signals are strong. These results suggest that inference errors stem from limited precision in perceiving signal strength. This mechanism offers a unified explanation for various belief anomalies and contributes to our understanding of how people respond to uncertainty. |
| Supervisor | Köster Mats |
| Department | Undergraduate Studies BA |
| Full text | https://www.etd.ceu.edu/2025/machavariani_giorgi.pdf |
Visit the CEU Library.
© 2007-2025, Central European University