Modelling the drivers of financial anxiety: a Bayesian network approach
Elvira Anna Graziano, Flaminia Musella, Gerardo PetroccionePurpose
This study aims to examine how financial anxiety is shaped within a system of interrelated cognitive, behavioural and contextual factors, with particular attention to the role of financial literacy and its relationship with financial behaviour and financial well-being.
Design/methodology/approach
A Bayesian network model is employed to analyse conditional independencies among financial literacy, financial behaviour, financial well-being and contextual stressors using survey data from a nationally representative sample of USA adults. Bayesian networks enable the estimation of direct and indirect relationships and support scenario-based probabilistic simulations.
Findings
The results show that financial anxiety emerges from interrelations among socio-demographic aspects rather than from isolated determinants. Financial literacy reduces the likelihood of extreme financial anxiety and improves financial well-being. The findings also highlight the relation between financial anxiety and financial well-being and the increasing role of contextual stressors.
Originality/value
This study contributes to the literature by conceptualising financial anxiety as an emergent system-level outcome and providing a theoretical reconciliation of mixed evidence on financial literacy. Methodologically, it demonstrates the value of Bayesian networks for modelling complex interdependencies and supporting probabilistic analysis in consumer finance research.