|
Learning Preferences from Response Times| title | Learning Preferences from Response Times |
|---|
| start_date | 2022/09/29 |
|---|
| schedule | 11h-12h |
|---|
| online | no |
|---|
| location_info | En ligne |
|---|
| summary | This talk reports on recent work showing how to use response time data and choice frequencies to elicit preferences nonparametrically when choices are stochastic. It discusses theoretical results and their empirical implementation. The talk also discusses two recent applications. The first shows how we can separate non-transitive preferences from noise-generated transitivity violations, a problem so far unresolved. The second shows how to apply the techniques to improve the analysis of survey data and determine whether a group (defined, e.g., by gender, age cohort, socioeconomic status, etc.) prefers an option over its alternative more than another group, even when individual choices are noisy, and without any assumptions on the noise |
|---|
| responsibles | Chassagnon, Apouey |
|---|
Workflow history| from state (1) | to state | comment | date |
|---|
| submitted | published | | 2022/11/22 16:22 UTC |
| |
|