Transition choice probabilities and welfare analysis in random utility models

We study the descriptive and the normative consequences of attribute changes in standard discrete choice models. For additive random utility models, we derive expressions for the transition choice probabilities for a change in the systematic utility. We then use these expressions to compute the CDF’s of the compensating variation conditional on the initial and on the final choices. The conditional moments of the compensating variation are obtained as a one-dimensional integral of the transition choice probabilities. We also provide a stochastic version of Shephard’s Lemma when transitions are observed. Example of the logit and the disaggregated CES are also studied.