Changing with the tide: Semiparametric estimation of preference dynamics

We contrast the discovered preference hypothesis against the theory of coherent arbitrariness in a split-sample stated choice experiment on flood-risk exposure. A semiparametric local multinomial logit model is developed as an alternative to the Swait and Louviere (1993) test procedure controlling for preference dynamics within and between samples. The proposed model supports the discovered preference hypothesis by means of a decaying starting point bias. The Swait and Louviere (1993) test procedure reaches a different conclusion. It rejects the assumption of stable preferences, but most preference dynamics tend to be smoothed out, causing a more erratic pattern of preference dynamics.