- This event has passed.
Giovanni COMPIANI (University of Chicago Booth School of Business) – “A Method to Estimate Discrete Choice Models that is Robust to Consumer Search”
Time: 3:00pm -4:15pm
Date: 14th of September 2021
Room : en visio
Giovanni COMPIANI (University of Chicago Booth School of Business) – “A Method to Estimate Discrete Choice Models that is Robust to Consumer Search”
Abstract: We state sufficient conditions under which choice data alone suffices to identify preferences when consumers are not fully informed about attributes of goods. Canonical models will be biased: the value of hidden attributes will be understated because consumers will be unresponsive to some variation in those attributes. In our baseline case, consumers search goods in order of the component of utility observable to them without search. Under our assumptions, an alternative method of recovering preferences using cross derivatives of choice probabilities succeeds under both full information and a range of search models and is thus robust to what consumers know when they choose. Our approach can be used to recover preferences from choices made by imperfectly informed consumers, to test for full information, and to forecast how consumers will respond to information. We verify in a lab experiment that our approach succeeds in forecasting the response to new information and assessing the value of that information when consumers engage in costly search.
Organizers:
Benoît SCHMUTZ (Pôle d’économie du CREST)
Anthony STRITTMATTER (Pôle d’économie du CREST)
Sponsors:
CREST
Time: 3:00pm -4:15pm
Date: 14th of September 2021
Room : en visio
Giovanni COMPIANI (University of Chicago Booth School of Business) – “A Method to Estimate Discrete Choice Models that is Robust to Consumer Search”
Abstract: We state sufficient conditions under which choice data alone suffices to identify preferences when consumers are not fully informed about attributes of goods. Canonical models will be biased: the value of hidden attributes will be understated because consumers will be unresponsive to some variation in those attributes. In our baseline case, consumers search goods in order of the component of utility observable to them without search. Under our assumptions, an alternative method of recovering preferences using cross derivatives of choice probabilities succeeds under both full information and a range of search models and is thus robust to what consumers know when they choose. Our approach can be used to recover preferences from choices made by imperfectly informed consumers, to test for full information, and to forecast how consumers will respond to information. We verify in a lab experiment that our approach succeeds in forecasting the response to new information and assessing the value of that information when consumers engage in costly search.
Organizers:
Benoît SCHMUTZ (Pôle d’économie du CREST)
Anthony STRITTMATTER (Pôle d’économie du CREST)
Sponsors:
CREST