{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/quantification-under-prior-probability-shift","title":"Quantification under prior probability shift: the ratio estimator and its extensions","arxiv_id":"1807.03929","date":"2018-07-11","proceeding":null,"authors":["Afonso Fernandes Vaz","Rafael Izbicki","Rafael Bassi Stern"],"abstract":"The quantification problem consists of determining the prevalence of a given\nlabel in a target population. However, one often has access to the labels in a\nsample from the training population but not in the target population. A common\nassumption in this situation is that of prior probability shift, that is, once\nthe labels are known, the distribution of the features is the same in the\ntraining and target populations. In this paper, we derive a new lower bound for\nthe risk of the quantification problem under the prior shift assumption.\nComplementing this lower bound, we present a new approximately minimax class of\nestimators, ratio estimators, which generalize several previous proposals in\nthe literature. Using a weaker version of the prior shift assumption, which can\nbe tested, we show that ratio estimators can be used to build confidence\nintervals for the quantification problem. We also extend the ratio estimator so\nthat it can: (i) incorporate labels from the target population, when they are\navailable and (ii) estimate how the prevalence of positive labels varies\naccording to a function of certain covariates.","url_abs":"http://arxiv.org/abs/1807.03929v2","url_pdf":"http://arxiv.org/pdf/1807.03929v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"quantification-under-prior-probability-shift","repo_url":"https://github.com/afonsofvaz/ratio_estimator","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}