eprintid: 47539 rev_number: 19 eprint_status: archive userid: 1482 importid: 105 dir: disk0/00/04/75/39 datestamp: 2023-04-07 14:11:21 lastmod: 2023-08-30 13:10:13 status_changed: 2023-06-29 12:37:14 type: monograph metadata_visibility: show creators_name: Cavaillé, Charlotte creators_name: Van Der Straeten, Karine creators_name: Chen, Daniel L. creators_idrefppn: 241586615 creators_idrefppn: 119884348 creators_idrefppn: 241586631 creators_affiliation: Toulouse School of Economics; Institute for Advanced Study in Toulouse; CNRS creators_affiliation: Toulouse School of Economics; Institute for Advanced Study in Toulouse; CNRS creators_halaffid: 1002422;441569 creators_halaffid: 1002422;441569 title: Willingness to say? Optimal survey design for prediction ispublished: pub subjects: subjects_ECO abstract: Survey design often approximates a prediction problem: the goal is to select instruments that best predict the value of an unobserved construct or a future outcome. We demonstrate how advances in machine learning techniques can help choose among competing instruments. First, we randomly assign respondents to one of four survey instruments to predict a behavior defined by our validation strategy. Next, we assess the optimal instrument in two stages. A machine learning model first predicts the behavior using individual covariates and survey responses. Then, using doubly robust welfare maximization and prediction error from the first stage, we learn the optimal survey method and examine how it varies across education levels. date: 2023-03 date_type: published publisher: TSE Working Paper official_url: http://tse-fr.eu/pub/128022 faculty: tse divisions: tse language: en has_fulltext: TRUE view_date_year: 2023 full_text_status: public monograph_type: working_paper series: TSE Working Paper volume: 23-1424 place_of_pub: Toulouse pages: 23 institution: Université Toulouse 1 Capitole department: Toulouse School of Economics book_title: TSE Working Paper oai_identifier: oai:tse-fr.eu:128022 harvester_local_overwrite: department harvester_local_overwrite: creators_name harvester_local_overwrite: pending harvester_local_overwrite: creators_idrefppn harvester_local_overwrite: creators_halaffid harvester_local_overwrite: institution harvester_local_overwrite: place_of_pub harvester_local_overwrite: pages harvester_local_overwrite: hal_id harvester_local_overwrite: hal_version harvester_local_overwrite: hal_url harvester_local_overwrite: hal_passwd harvester_local_overwrite: publish_to_hal harvester_local_overwrite: title oai_lastmod: 2023-06-19T12:51:18Z oai_set: tse site: ut1 publish_to_hal: FALSE hal_id: hal-04062637 hal_passwd: 55s?k7 hal_version: 1 hal_url: https://hal.science/hal-04062637 citation: Cavaillé, Charlotte , Van Der Straeten, Karine and Chen, Daniel L. (2023) Willingness to say? Optimal survey design for prediction. TSE Working Paper, n. 23-1424, Toulouse document_url: https://publications.ut-capitole.fr/id/eprint/47539/1/wp_tse_1424.pdf