How do individuals respond to DCEs? Alternatives to utility maximisation

Health decisions are often made in a context of urgency and uncertainty, making other concepts, such as regret, also relevant. The random regret minimisation (RRM) model has been developed as an alternative to RUM. However the link between RRM and the psychological concept of regret remains unclear. In this study we tackle this issue by comparing the performance of the RUM and RRM models in the context of cancer follow-up care decisions. We develop a variant of the RRM mode. We verify the properties of this new model, random gain-loss maximisation (RGLM), in a simulation study. The comparison of RUM and RGLM models shows that RGLM allows for a small but significant improvement in in-sample fit, but similar level of predictive validity (out-of-sample fit). Initial anlysis suggests limited support for regret minimisation in the analysis of cancer follow-up care decisions. 

HERU researchers involved in this research project: Verity Watson and Mandy Ryan

External collaborators: N. Krucien (Evidera)


Krucien, N.Ryan, M. and Watson, V. (2014) 'What can hypothetical choices tell us about unobservable behaviours? Regret minimisation, rational non-attendance and randomness in choice experiment responses', Health Economics in the Age of Longevity: a Joint iHEA & ECHE Congress, Trinity College, Dublin, 13-16 July 2014.

Krucien, N., Watson, V. and Ryan, M. (2015) 'Is there room for regret minimisation in patients’ decisions for cancer-related procedures? Evidence from three discrete choice experiments', Health Economists' Study Group (HESG) Summer 2015, Lancaster University, Lancaster, 22-24 June 2015.