Regret Aversion and Dynamic Choice Stephen Lovelady This research looks into the possibility of bringing together two distinct and empirically successful areas of behavioural economics; regret aversion and quasi-hyperbolic discounting. Standard regret aversion theory (Loomes and Sugden 1982, Bell 1982) operates in a world of uncertainty, where regret aversion arises due to the possibility of, having made an initial choice, a more preferable option arising when uncertainty is resolved. However, the common intuition in the literature behind the psychological notion of “regret” does not explicitly mention uncertainty. Merely, ...regret is the result of comparing one's outcome with a better outcome that would have occurred had a different alternative been selected (Tsiros and Mittal, 2000, p.402) Hence, my question is whether anticipation of regret can arise and have predictive power in a wider range of circumstances, such as certain worlds and dynamic models. This concept of “preference reversal” creating the possibility of regret naturally lends itself to the framework of quasi-hyperbolic discounting. There is a clear understanding that this method of discounting can lead to the supposed problem of “time inconsistency”, where preferences can change over time simply due to the present bias of the discounting model. It is precisely this inconsistency, and the preference reversal it can generate, that I seek to exploit, by combining it with psychological effects, such as regret, that this preference reversal will cause. The work seeks to answer the question of whether these two types of preference reversal are linked. The quasi-hyperbolic framework, however, is still in a stage of infancy and it will require much work and adaptation to mould it into a partner framework for the regret aversion model. This principally surrounds the extension of the most common discrete time model to a more continuous time model, and I hope to use the joint regret/present bias model I generate to shed new light on the notion of the “present bias” in a continuous setting to enable future theoretical and empirical work.