This paper provides a new approach to understanding bankers' risk-taking behavior. We build upon prior studies that suggest artificial intelligence algorithms are an effective approach to obtaining this understanding. Our approach uses behavioral finance and a unique decision-making model. Although the decision-making literature is replete with descriptions and explanations of creditors and investors' perceptions and judgment, it does not provide an algorithmic model that incorporates a more flexible approach to how creditors subjectively valuate risky projects. Specifically, a model is presented where 33 corporate bankers realized ex ante that they were unable to accurately model the underlying uncertainty that characterizes a company's need for a loan. The results indicate that bankers' risk assessments result in different evaluations of financial information regarding loans. This approach depicts an integrative algorithmic modelling process, whereby limits in the amount of historical conditional information prohibit the use of more complex econometric techniques.
Rodgers, W., Hudson, R., & Economou, F. (2023). Modelling credit and investment decisions based on AI algorithmic behavioral pathways. Technological Forecasting and Social Change, 191, Article 122471. https://doi.org/10.1016/j.techfore.2023.122471