Estimating discrete choice experiments : theoretical fundamentals

This working paper overviews theoretical foundations and estimators derived from econometric models used to analyze stated choices proposed in Discrete Choice Experiment (DCE)surveys. Discrete Choice Modelling is adapted to the case where the variable to be explained is a qualitative variable which cannot be ranked in relation to each other. A situation which occurs in many cases in everyday life as people often have to choose only one alternative among a proposed set of different ones in many fields (early in the morning, just think about how you pick clothes for instance). DCE is a Stated Preference method in which preferences are elicited through repeated fictional choices, proposed to a sample of respondents. Compared to RevealedPreference methods, DCEs allow for an ex ante evaluation of public policies that do not yet exists.