Data Engineering for Cognitive Economics

This is the authors' abstract. We don't add key points for this paper.

Cognitive economics studies imperfect information and decision-making mistakes. A central scientific challenge is that these can’t be identified in standard choice data. Overcoming this challenge calls for data engineering, in which new data forms are introduced to separately identify preferences, beliefs, and other model constructs. I present applications to traditional areas of economic research, such as wealth accumulation, earnings, and consumer spending. I also present less traditional applications to assessment of decision-making skills, and to human–AI interactions. Methods apply both to individual and to collective decisions. I make the case for broader application of data engineering beyond cognitive economics. It allows symbiotic advances in modeling and measurement. It cuts across existing boundaries between disciplines and styles of research. (JEL C45, C80, D15, D80, D91, G50, J24)