Mapping Financial Mindsets: A Two-Stage Unsupervised Framework for Behavioral Profiling Using High-Dimensional Psychometric Data

Financial well-being reflects a complex interplay of psychological traits, behaviors, andknowledge, yet most segmentation approaches rely on limited indicators or assume lineardependencies. This paper proposes a novel framework for identifying latent profiles of financialmindset and behavior by modeling nonlinear interactions across multiple psychometric domains.Using data from a proprietary assessment (N = 144) and the nationally representative U.S. CFPBFinancial Well-Being Survey (N = 6,233), we uncover recurring psychologically groundedprofiles that generalize across populations and survey formats. These profiles show stronginternal structure and are significantly associated with life satisfaction, psychological well-being,and financial health. By capturing deeper drivers of financial behavior, our approach movesbeyond conventional demographic or transaction-based segmentation. The resulting framework isreplicable, scalable, and well suited to behavioral finance applications, including financialadvising, wellness interventions, and the design of personalized tools and communications aimedat improving financial decision-making and well-being.