{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T07:19:01Z","timestamp":1791875941727,"version":"4.3.4"},"reference-count":76,"publisher":"The Econometric Society","issue":"2","license":[{"start":{"date-parts":[[2023,1,1]],"date-time":"2023-01-01T00:00:00Z","timestamp":1672531200000},"content-version":"unspecified","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["QE"],"abstract":"<jats:p>This paper provides estimation and inference methods for conditional average treatment effects (CATE) characterized by a high\u2010dimensional parameter in both homogeneous cross\u2010sectional and unit\u2010heterogeneous dynamic panel data settings. In our leading example, we model CATE by interacting the base treatment variable with explanatory variables. The first step of our procedure is orthogonalization, where we partial out the controls and unit effects from the outcome and the base treatment and take the cross\u2010fitted residuals. This step uses a novel generic cross\u2010fitting method that we design for weakly dependent time series and panel data. This method \u201cleaves out the neighbors\u201d when fitting nuisance components, and we theoretically power it by using Strassen's coupling. As a result, we can rely on any modern machine learning method in the first step, provided it learns the residuals well enough. Second, we construct an orthogonal (or residual) learner of CATE\u2014the lasso CATE\u2014that regresses the outcome residual on the vector of interactions of the residualized treatment with explanatory variables. If the complexity of CATE function is simpler than that of the first\u2010stage regression, the orthogonal learner converges faster than the single\u2010stage regression\u2010based learner. Third, we perform simultaneous inference on parameters of the CATE function using debiasing. We also can use ordinary least squares in the last two steps when CATE is low\u2010dimensional. In heterogeneous panel data settings, we model the unobserved unit heterogeneity as a weakly sparse deviation from Mundlak's (1978) model of correlated unit effects as a linear function of time\u2010invariant covariates and make use of L1\u2010penalization to estimate these models. We demonstrate our methods by estimating price elasticities of groceries based on scanner data. We note that our results are new even for the cross\u2010sectional (i.i.d.) case.<\/jats:p>","DOI":"10.3982\/qe1670","type":"journal-article","created":{"date-parts":[[2023,5,11]],"date-time":"2023-05-11T09:06:37Z","timestamp":1683795997000},"page":"471-510","source":"Crossref","is-referenced-by-count":35,"title":["Inference on heterogeneous treatment effects in high\u2010dimensional dynamic panels under weak dependence"],"prefix":"10.3982","volume":"14","author":[{"given":"Vira","family":"Semenova","sequence":"first","affiliation":[{"name":"Department of Economics, University of California, Berkeley"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Matt","family":"Goldman","sequence":"additional","affiliation":[{"name":"Meta"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Victor","family":"Chernozhukov","sequence":"additional","affiliation":[{"name":"Department of Economics, MIT"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Matt","family":"Taddy","sequence":"additional","affiliation":[{"name":"Amazon"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"2359","published-online":{"date-parts":[[2023]]},"reference":[{"key":"quan200275-bib-0001","doi-asserted-by":"crossref","first-page":"277","DOI":"10.2307\/2297968","volume":"58","author":"Arellano","year":"1991","journal-title":"Review of Economic Studies"},{"key":"quan200275-bib-0002","doi-asserted-by":"crossref","first-page":"7353","DOI":"10.1073\/pnas.1510489113","volume":"113","author":"Athey","year":"2016","journal-title":"Proceedings of the National Academy of Sciences"},{"key":"quan200275-bib-0003","doi-asserted-by":"crossref","unstructured":"Banerjee, Abhijit, Arun G. Chandrasekhar, Suresh Dalpath, Esther Duflo, John Floretta, Matthew O. Jackson, Harini Kannan, Francine Loza, Anirudh Sankar, Anna Schrimpf, and Maheshwor Shrestha (2021), \u201cSelecting the most effective nudge: Evidence from a large-scale experiment on immunization.\u201d Available at https:\/\/arxiv.org\/abs\/2104.09645.","DOI":"10.3386\/w28726"},{"key":"quan200275-bib-0004","doi-asserted-by":"publisher","DOI":"10.1093\/biomet\/asu056"},{"key":"quan200275-bib-0005","doi-asserted-by":"crossref","first-page":"2369","DOI":"10.3982\/ECTA9626","volume":"80","author":"Belloni","year":"2012","journal-title":"Econometrica"},{"key":"quan200275-bib-0006","doi-asserted-by":"crossref","first-page":"521","DOI":"10.3150\/11-BEJ410","volume":"19","author":"Belloni","year":"2013","journal-title":"Bernoulli"},{"key":"quan200275-bib-0007","first-page":"3643","volume":"46","author":"Belloni","year":"2019","journal-title":"Annals of Statistics"},{"key":"quan200275-bib-0008","doi-asserted-by":"crossref","first-page":"233","DOI":"10.3982\/ECTA12723","volume":"85","author":"Belloni","year":"2017","journal-title":"Econometrica"},{"key":"quan200275-bib-0009","doi-asserted-by":"crossref","unstructured":"Belloni, Alexandre, Victor Chernozhukov, and Christian Hansen (2010), \u201cLasso methods for Gaussian instrumental variables models.\u201d arXiv preprint, arXiv:1012.1297.","DOI":"10.2139\/ssrn.1908409"},{"key":"quan200275-bib-0010","doi-asserted-by":"crossref","unstructured":"Belloni, Alexandre, Victor Chernozhukov, and Christian Hansen (2011), \u201cInference for high-dimensional sparse econometric models.\u201d arXiv preprint, arXiv:1201.0220.","DOI":"10.1920\/wp.cem.2011.4111"},{"key":"quan200275-bib-0011","doi-asserted-by":"crossref","first-page":"608","DOI":"10.1093\/restud\/rdt044","volume":"81","author":"Belloni","year":"2014","journal-title":"Review of Economic Studies"},{"key":"quan200275-bib-0012","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1093\/biomet\/asu056","volume":"102","author":"Belloni","year":"2014","journal-title":"Biometrika"},{"key":"quan200275-bib-0013","doi-asserted-by":"crossref","first-page":"606","DOI":"10.1080\/07350015.2016.1166116","volume":"34","author":"Belloni","year":"2016","journal-title":"Journal of Business & Economic Statistics"},{"key":"quan200275-bib-0014","doi-asserted-by":"crossref","first-page":"255","DOI":"10.1007\/BF00569992","volume":"74","author":"Berbee","year":"1987","journal-title":"Probability Theory and Related Fields"},{"key":"quan200275-bib-0015","unstructured":"Bonhomme, Stephane, Thibaut Lamadon, and Elena Manresa (2019a), \u201cDiscretizing unobserved heterogeneity.\u201d"},{"key":"quan200275-bib-0016","doi-asserted-by":"crossref","first-page":"699","DOI":"10.3982\/ECTA15722","volume":"87","author":"Bonhomme","year":"2019","journal-title":"Econometrica"},{"key":"quan200275-bib-0017","doi-asserted-by":"crossref","first-page":"594","DOI":"10.1198\/jasa.2011.tm10155","volume":"106","author":"Cai","year":"2011","journal-title":"Journal of the American Statistical Association"},{"key":"quan200275-bib-0018","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1016\/0304-4076(82)90094-X","volume":"18","author":"Chamberlain","year":"1982","journal-title":"Journal of Econometrics"},{"key":"quan200275-bib-0019","author":"Chen","year":"2020","journal-title":"Journal of Econometrics"},{"key":"quan200275-bib-0020","doi-asserted-by":"crossref","first-page":"C1","DOI":"10.1111\/ectj.12097","volume":"21","author":"Chernozhukov","year":"2018","journal-title":"Econometrics Journal"},{"key":"quan200275-bib-0021","doi-asserted-by":"crossref","first-page":"2786","DOI":"10.1214\/13-AOS1161","volume":"41","author":"Chernozhukov","year":"2013","journal-title":"Annals of Statistics"},{"key":"quan200275-bib-0022","first-page":"1564","volume":"42","author":"Chernozhukov","year":"2014","journal-title":"Annals of Statistics"},{"key":"quan200275-bib-0023","doi-asserted-by":"crossref","first-page":"2309","DOI":"10.1214\/16-AOP1113","volume":"45","author":"Chernozhukov","year":"2017","journal-title":"Annals of Probability"},{"key":"quan200275-bib-0024","doi-asserted-by":"crossref","first-page":"1867","DOI":"10.1093\/restud\/rdy065","volume":"86","author":"Chernozhukov","year":"2019","journal-title":"Review of Economic Studies"},{"key":"quan200275-bib-0025","doi-asserted-by":"crossref","unstructured":"Chernozhukov, Victor, Mert Demirer, Esther Duflo, and Iv\u00e1n Fern\u00e1ndez-Val (2017), \u201cGeneric machine learning inference on heterogenous treatment effects in randomized experiments.\u201d arXiv e-prints, arXiv:1712.04802.","DOI":"10.3386\/w24678"},{"key":"quan200275-bib-0026","unstructured":"Chernozhukov, Victor, Chris Hansen, and Martin Spindler (2016), \u201cHigh-dimensional metrics in R.\u201d Available at https:\/\/arxiv.org\/abs\/1603.01700."},{"key":"quan200275-bib-0027","doi-asserted-by":"crossref","first-page":"649","DOI":"10.1146\/annurev-economics-012315-015826","volume":"7","author":"Chernozhukov","year":"2015","journal-title":"Annual Review of Economics"},{"key":"quan200275-bib-0028","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1257\/000282803321455142","volume":"93","author":"Chevalier","year":"2003","journal-title":"American Economic Review"},{"key":"quan200275-bib-0029","unstructured":"Chiang, Harold D. (2018), \u201cMany average partial effects: With an application to text regression.\u201d arXiv e-prints, arXiv:1812.09397."},{"key":"quan200275-bib-0030","unstructured":"Chiang, Harold D., Kengo Kato, Yukun Ma, and Yuya Sasaki (2019), \u201cMultiway cluster robust double\/debiased machine learning.\u201d arXiv e-prints, arXiv:1909.03489."},{"key":"quan200275-bib-0031","unstructured":"Colangelo, Kyle and Ying-Ying Lee (2020), \u201cDouble debiased machine learning nonparametric inference with continuous treatments.\u201d arXiv e-prints, arXiv:2004.03036."},{"key":"quan200275-bib-0032","doi-asserted-by":"publisher","DOI":"10.1162\/rest_a_00850"},{"key":"quan200275-bib-0033","doi-asserted-by":"crossref","first-page":"509","DOI":"10.1007\/BF00534202","volume":"62","author":"Dudley","year":"1983","journal-title":"Zeitschrift f\u00fcr Wahrscheinlichkeitstheorie und verwandte Gebiete"},{"key":"quan200275-bib-0034","unstructured":"Fan, Qingliang, Yu-Chin Hsu, Robert P. Lieli, and Yichong Zhang (2019), \u201cEstimation of conditional average treatment effects with high-dimensional data.\u201d arXiv e-prints, arXiv:1908.02399."},{"key":"quan200275-bib-0035","doi-asserted-by":"crossref","first-page":"181","DOI":"10.3982\/ECTA16901","volume":"89","author":"Farrell","year":"2021","journal-title":"Econometrica"},{"key":"quan200275-bib-0036","doi-asserted-by":"crossref","first-page":"453","DOI":"10.3982\/QE75","volume":"4","author":"Fernandez-Val","year":"2013","journal-title":"Quantitative Economics"},{"key":"quan200275-bib-0037","doi-asserted-by":"crossref","first-page":"109","DOI":"10.1146\/annurev-economics-080217-053542","volume":"10","author":"Fernandez-Val","year":"2018","journal-title":"Annual Review of Economics"},{"key":"quan200275-bib-0038","doi-asserted-by":"crossref","unstructured":"Gao, Wayne Yuan, and Ming Li (2019), \u201cRobust semiparametric estimation in panel multinomial choice models.\u201d","DOI":"10.2139\/ssrn.3282293"},{"key":"quan200275-bib-0039","doi-asserted-by":"crossref","first-page":"1","DOI":"10.2307\/1906935","volume":"12","author":"Haavelmo","year":"1944","journal-title":"Econometrica"},{"key":"quan200275-bib-0040","doi-asserted-by":"crossref","first-page":"1152","DOI":"10.1017\/S0266466611000028","volume":"27","author":"Hahn","year":"2011","journal-title":"Econometric Theory"},{"key":"quan200275-bib-0041","volume-title":"Proceedings of the Second Prague Symposium on Asymptotic Statistics","author":"Hasminskii","year":"1979"},{"key":"quan200275-bib-0042","unstructured":"Jacob, Daniel, Wolfgang Karl H\u00e4rdle, and Stefan Lessmann (2019), \u201cGroup average treatment effects for observational studies.\u201d"},{"key":"quan200275-bib-0043","first-page":"2869","volume":"2","author":"Javanmard","year":"2014","journal-title":"Journal of Machine Learning Research"},{"key":"quan200275-bib-0044","unstructured":"Kennedy, Edward, (2020), \u201cTowards optimal doubly robust estimation of heterogeneous causal effects.\u201d arXiv preprint, arXiv:2004.14497."},{"key":"quan200275-bib-0045","unstructured":"Klosin, Sylvia (2021), \u201cAutomatic double machine learning for continuous treatment effects.\u201d"},{"key":"quan200275-bib-0046","doi-asserted-by":"crossref","first-page":"243","DOI":"10.1017\/S0266466615000304","volume":"32","author":"Kock","year":"2016","journal-title":"Econometric Theory"},{"key":"quan200275-bib-0047","doi-asserted-by":"crossref","first-page":"71","DOI":"10.1016\/j.jeconom.2016.06.001","volume":"195","author":"Kock","year":"2016","journal-title":"Journal of Econometrics"},{"key":"quan200275-bib-0048","doi-asserted-by":"crossref","first-page":"295","DOI":"10.1017\/S0266466618000087","volume":"35","author":"Kock","year":"2019","journal-title":"Econometric Theory"},{"key":"quan200275-bib-0049","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1017\/S0266466605050036","volume":"21","author":"Leeb","year":"2005","journal-title":"Econometric Theory"},{"key":"quan200275-bib-0050","doi-asserted-by":"crossref","first-page":"148","DOI":"10.1016\/j.jeconom.2015.09.005","volume":"190","author":"Lu","year":"2016","journal-title":"Journal of Econometrics"},{"key":"quan200275-bib-0051","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1016\/j.jeconom.2019.07.008","volume":"215","author":"Lu","year":"2020","journal-title":"Journal of Econometrics"},{"key":"quan200275-bib-0052","unstructured":"Luo, Ye and Martin Spindler (2016), \u201cHigh-dimensional L2-boosting: Rate of convergence.\u201d arXiv e-prints, arXiv:1602.08927."},{"key":"quan200275-bib-0053","unstructured":"Manresa, Elena (2016), \u201cEstimating the structure of social interactions using panel data.\u201d"},{"key":"quan200275-bib-0054","doi-asserted-by":"crossref","first-page":"613","DOI":"10.1016\/j.jeconom.2018.06.016","volume":"206","author":"Moon","year":"2018","journal-title":"Journal of Econometrics"},{"key":"quan200275-bib-0055","doi-asserted-by":"crossref","first-page":"69","DOI":"10.2307\/1913646","volume":"46","author":"Mundlak","year":"1978","journal-title":"Econometrica"},{"key":"quan200275-bib-0056","doi-asserted-by":"crossref","first-page":"538","DOI":"10.1214\/12-STS400","volume":"27","author":"Negahban","year":"2012","journal-title":"Statistical Science"},{"key":"quan200275-bib-0057","doi-asserted-by":"crossref","first-page":"1824","DOI":"10.2307\/1911408","volume":"49","author":"Nickell","year":"1981","journal-title":"Econometrica"},{"key":"quan200275-bib-0058","doi-asserted-by":"crossref","unstructured":"Nie, Xinkun and Stefan Wager (2020), \u201cQuasi-oracle estimation of heterogeneous treatment effects.\u201d Biometrika.","DOI":"10.1093\/biomet\/asaa076"},{"key":"quan200275-bib-0059","unstructured":"Oprescu, Miruna, Vasilis Syrgkanis, and Zhiwei Steven Wu (2018), \u201cOrthogonal random forest for causal inference.\u201d arXiv e-prints, arXiv:1806.03467."},{"key":"quan200275-bib-0060","doi-asserted-by":"crossref","first-page":"931","DOI":"10.2307\/1912705","volume":"56","author":"Robinson","year":"1988","journal-title":"Econometrica"},{"key":"quan200275-bib-0061","doi-asserted-by":"crossref","first-page":"1139","DOI":"10.1214\/aos\/1176350055","volume":"14","author":"Schick","year":"1986","journal-title":"Annals of Statistics"},{"key":"quan200275-bib-0062","unstructured":"Schmidt-Hieber, Johannes (2017), \u201cNonparametric regression using deep neural networks with ReLU activation function.\u201d arXiv e-prints, arXiv:1708.06633."},{"issue":"2","key":"quan200275-bib-0063","doi-asserted-by":"crossref","first-page":"264","DOI":"10.1093\/ectj\/utaa027","volume":"24","author":"Semenova","year":"2021","journal-title":"Econometrics Journal"},{"key":"quan200275-bib-0064","doi-asserted-by":"crossref","unstructured":"Semenova, Vira, Matt Goldman, Victor Chernozhukov, and Matt Taddy (2023), \u201cSupplement to \u2018Inference on heterogeneous treatment effects in high-dimensional dynamic panels under weak dependence\u2019.\u201d Quantitative Economics Supplemental Material, 14, https:\/\/doi.org\/10.3982\/QE1670.","DOI":"10.3982\/QE1670"},{"key":"quan200275-bib-0065","doi-asserted-by":"crossref","first-page":"423","DOI":"10.1214\/aoms\/1177700153","volume":"36","author":"Strassen","year":"1965","journal-title":"Annals of Mathematical Statistics"},{"key":"quan200275-bib-0066","doi-asserted-by":"crossref","first-page":"1824","DOI":"10.3982\/ECTA12560","volume":"84","author":"Su","year":"2016","journal-title":"Econometrica"},{"key":"quan200275-bib-0067","unstructured":"Syrganis, Vasilis and Manolis Zampetakis (2020), \u201cEstimation and inference with trees and forests in high dimensions.\u201d arXiv e-prints, arXiv:2007.03210."},{"key":"quan200275-bib-0068","doi-asserted-by":"crossref","first-page":"1335","DOI":"10.3982\/QE886","volume":"9","author":"Ura","year":"2018","journal-title":"Quantitative Economics"},{"key":"quan200275-bib-0069","first-page":"1166","volume":"42","author":"Geer","year":"2014","journal-title":"Annals of Statistics"},{"key":"quan200275-bib-0070","volume-title":"Topics in Optimal Transportation","volume":"58","author":"Villani","year":"2007"},{"key":"quan200275-bib-0071","doi-asserted-by":"crossref","first-page":"1228","DOI":"10.1080\/01621459.2017.1319839","volume":"113","author":"Wager","year":"2018","journal-title":"Journal of the American Statistical Association"},{"key":"quan200275-bib-0072","unstructured":"Wager, Stefan and Guenther Walther (2015), \u201cAdaptive concentration of regression trees, with application to random forests.\u201d arXiv e-prints, arXiv:1503.06388."},{"key":"quan200275-bib-0073","doi-asserted-by":"crossref","first-page":"217","DOI":"10.1111\/rssb.12026","volume":"76","author":"Zhang","year":"2014","journal-title":"Journal of the Royal Statistical Society: Series B (Statistical Methodology)"},{"key":"quan200275-bib-0074","doi-asserted-by":"crossref","first-page":"1895","DOI":"10.1214\/16-AOS1512","volume":"45","author":"Zhang","year":"2017","journal-title":"Annals of Statistics"},{"key":"quan200275-bib-0075","doi-asserted-by":"crossref","unstructured":"Zheng, Wenjing and Mark J. van der Laan (2010), \u201cAsymptotic theory for cross-validated targeted maximum likelihood estimation.\u201d Technical report, UC Berkeley Division of Biostatistics.","DOI":"10.2202\/1557-4679.1181"},{"key":"quan200275-bib-0076","unstructured":"Zimmert, Michael and Michael Lechner (2019), \u201cNonparametric estimation of causal heterogeneity under high-dimensional confounding.\u201d arXiv e-prints, arXiv:1908.08779."}],"container-title":["Quantitative Economics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.3982\/QE1670","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,5,11]],"date-time":"2023-05-11T09:08:45Z","timestamp":1683796125000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.econometricsociety.org\/doi\/10.3982\/QE1670"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023]]},"references-count":76,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2023]]}},"alternative-id":["10.3982\/QE1670"],"URL":"https:\/\/doi.org\/10.3982\/qe1670","relation":{},"ISSN":["1759-7323"],"issn-type":[{"value":"1759-7323","type":"print"}],"subject":[],"published":{"date-parts":[[2023]]}}}