We present a novel model for a time series of individual binary decisions which depends on the history of prices. The model is based on the Bayesian learning procedure which is at the core of sequential decision making. We show that the model capture dependence on past events and past priors in a straightforward fashion, the model capture some dependence on initial condition, here in the form of the prior at the start of the decision period, and that estimation through maximum likelihood is straightforward. We estimate the parameters of the model on a sample of Swedish households who have to decide over time between competing electricity contracts. The estimated parameters suggest that households respond to prices by switching between contracts, and that the response can be rather substantial for alternative price processes