Sequential Decision Making in Dynamic Environments

This research article presents a novel experimental and econometric approach for identifying decision strategies in sequential decision-making contexts. The authors, Evgeny Kagan, Stephen Leider, and Ozge Sahin, outline the complexities involved in characterizing behavior in sequential problems, particularly due to multiple decision rules that may overlap. The study introduces a method that elicits conditional strategies and measures policy adherence through maximum likelihood estimation. By applying this approach to various sequential problems, the findings indicate that decision-makers often respond less strongly to the dynamic aspects of problems with stopping structures. Additionally, the format of decision elicitation significantly impacts policy adoption. The implications of these findings are particularly relevant for firms that serve customers making sequential decisions, as failing to account for relevant decision rules can lead to substantial profit losses. The article contributes to the understanding of dynamic decision-making and offers a framework that enhances the identification of decision policies.