Examining predictive models among discounting models

Abstract

Both descriptive and normative arguments claim that the discount rate to be applied to public projects should be elicited from individual intertemporal preferences. We present a methodology to analyze data from experimental surveys on intertemporal preferences. Focusing on the exponential and the hyperbolic discounting models, we model the experimental data published by Thaler (1981) by means of different specifications. Standard measures of goodness of prediction are then applied to fitted data to select among alternative specifications. We first present our approach by applying it to simulated data. We then present a procedure for statistical estimation of the sample discount rate, testing four specifications. The estimation procedure we adopt can be easily extended to other functional forms, allowing for non-linearity of the valuation function, inclusion of socio-economic individual characteristics as regressors, and different specifications for the discounting model.