{"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/difference-in-differences-with-multiple-time","title":"Difference-in-Differences with Multiple Time Periods","arxiv_id":"1803.09015","date":"2020-08-18","proceeding":null,"authors":[],"abstract":"In this article, we consider identification, estimation, and inference\nprocedures for treatment effect parameters using Difference-in-Differences\n(DID) with (i) multiple time periods, (ii) variation in treatment timing, and\n(iii) when the \"parallel trends assumption\" holds potentially only after\nconditioning on observed covariates. We show that a family of causal effect\nparameters are identified in staggered DID setups, even if differences in\nobserved characteristics create non-parallel outcome dynamics between groups.\nOur identification results allow one to use outcome regression, inverse\nprobability weighting, or doubly-robust estimands. We also propose different\naggregation schemes that can be used to highlight treatment effect\nheterogeneity across different dimensions as well as to summarize the overall\neffect of participating in the treatment. We establish the asymptotic\nproperties of the proposed estimators and prove the validity of a\ncomputationally convenient bootstrap procedure to conduct asymptotically valid\nsimultaneous (instead of pointwise) inference. Finally, we illustrate the\nrelevance of our proposed tools by analyzing the effect of the minimum wage on\nteen employment from 2001--2007. Open-source software is available for\nimplementing the proposed methods.","url_abs":"http://arxiv.org/abs/1803.09015v3","url_pdf":"http://arxiv.org/pdf/1803.09015v3.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":"difference-in-differences-with-multiple-time","repo_url":"https://github.com/bcallaway11/did","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"difference-in-differences-with-multiple-time","repo_url":"https://github.com/bcallaway11/twfeweights","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":null,"task_name":"valid"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1803.09015","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}