{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,15]],"date-time":"2026-10-15T10:10:48Z","timestamp":1792059048557,"version":"4.4.0"},"reference-count":105,"publisher":"Wiley","issue":"5","license":[{"start":{"date-parts":[[2022,6,23]],"date-time":"2022-06-23T00:00:00Z","timestamp":1655942400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["J of Applied Econometrics"],"published-print":{"date-parts":[[2022,8]]},"abstract":"<jats:title>Summary<\/jats:title>\n                  <jats:p>\n                    We move beyond\n                    <jats:italic>Is Machine Learning Useful for Macroeconomic Forecasting?<\/jats:italic>\n                    by adding the\n                    <jats:italic>how<\/jats:italic>\n                    . The current forecasting literature has focused on matching specific variables and horizons with a particularly successful algorithm. To the contrary, we study the usefulness of the underlying features driving ML gains over standard macroeconometric methods. We distinguish four so\u2010called features (nonlinearities, regularization, cross\u2010validation, and alternative loss function) and study their behavior in both the data\u2010rich and data\u2010poor environments. To do so, we design experiments that allow to identify the \u201ctreatment\u201d effects of interest. We conclude that\n                    <jats:bold>(i)<\/jats:bold>\n                    nonlinearity is the true game changer for macroeconomic prediction,\n                    <jats:bold>(ii)<\/jats:bold>\n                    the standard factor model remains the best regularization,\n                    <jats:bold>(iii)<\/jats:bold>\n                    K\u2010fold cross\u2010validation is the best practice, and\n                    <jats:bold>(iv)<\/jats:bold>\n                    the is preferred to the \u2010insensitive in\u2010sample loss. The forecasting gains of nonlinear techniques are associated with high macroeconomic uncertainty, financial stress and housing bubble bursts. Furthermore, ML nonlinearities are helpful when considering density forecasts.\n                  <\/jats:p>","DOI":"10.1002\/jae.2910","type":"journal-article","created":{"date-parts":[[2022,5,13]],"date-time":"2022-05-13T21:38:28Z","timestamp":1652477908000},"page":"920-964","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":182,"title":["How is machine learning useful for macroeconomic forecasting?"],"prefix":"10.1002","volume":"37","author":[{"given":"Philippe","family":"Goulet Coulombe","sequence":"first","affiliation":[{"name":"D\u00e9partement des Sciences \u00e9conomiques Universit\u00e9 du Qu\u00e9bec \u00e0 Montr\u00e9al Montr\u00e9al Qu\u00e9bec Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Maxime","family":"Leroux","sequence":"additional","affiliation":[{"name":"D\u00e9partement des Sciences \u00e9conomiques Universit\u00e9 du Qu\u00e9bec \u00e0 Montr\u00e9al Montr\u00e9al Qu\u00e9bec Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dalibor","family":"Stevanovic","sequence":"additional","affiliation":[{"name":"D\u00e9partement des Sciences \u00e9conomiques Universit\u00e9 du Qu\u00e9bec \u00e0 Montr\u00e9al Montr\u00e9al Qu\u00e9bec Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"St\u00e9phane","family":"Surprenant","sequence":"additional","affiliation":[{"name":"D\u00e9partement des Sciences \u00e9conomiques Universit\u00e9 du Qu\u00e9bec \u00e0 Montr\u00e9al Montr\u00e9al Qu\u00e9bec Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2022,6,23]]},"reference":[{"key":"e_1_2_12_2_1","doi-asserted-by":"publisher","DOI":"10.1162\/rest_a_00812"},{"key":"e_1_2_12_3_1","doi-asserted-by":"publisher","DOI":"10.1257\/aer.20161923"},{"key":"e_1_2_12_4_1","doi-asserted-by":"publisher","DOI":"10.1080\/07474938.2010.481556"},{"issue":"1","key":"e_1_2_12_5_1","doi-asserted-by":"crossref","first-page":"65","DOI":"10.2478\/demo-2013-0004","article-title":"Prediction of time series by statistical learning: General losses and fast rates","volume":"1","author":"Alquier P.","year":"2013","journal-title":"Dependence Modeling"},{"key":"e_1_2_12_6_1","doi-asserted-by":"publisher","DOI":"10.3982\/ECTA10008"},{"key":"e_1_2_12_7_1","doi-asserted-by":"publisher","DOI":"10.7208\/chicago\/9780226613475.003.0021"},{"key":"e_1_2_12_8_1","doi-asserted-by":"publisher","DOI":"10.1214\/18-AOS1709"},{"key":"e_1_2_12_9_1","doi-asserted-by":"publisher","DOI":"10.21034\/qr.2511"},{"key":"e_1_2_12_10_1","doi-asserted-by":"publisher","DOI":"10.1093\/qje\/qjw024"},{"key":"e_1_2_12_11_1","doi-asserted-by":"publisher","DOI":"10.1093\/restud\/rdx008"},{"key":"e_1_2_12_12_1","doi-asserted-by":"publisher","DOI":"10.1257\/aer.20190789"},{"key":"e_1_2_12_13_1","doi-asserted-by":"publisher","DOI":"10.3982\/ECTA12723"},{"key":"e_1_2_12_14_1","doi-asserted-by":"publisher","DOI":"10.3982\/ECTA11085"},{"key":"e_1_2_12_15_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jedc.2013.03.007"},{"key":"e_1_2_12_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ins.2011.12.028"},{"key":"e_1_2_12_17_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.csda.2017.11.003"},{"key":"e_1_2_12_18_1","doi-asserted-by":"publisher","DOI":"10.3982\/ECTA6248"},{"key":"e_1_2_12_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2005.01.027"},{"key":"e_1_2_12_20_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2019.07.007"},{"key":"e_1_2_12_21_1","doi-asserted-by":"publisher","DOI":"10.1111\/1468-0289.665"},{"key":"e_1_2_12_22_1","doi-asserted-by":"publisher","DOI":"10.1023\/A:1010933404324"},{"key":"e_1_2_12_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijforecast.2019.02.007"},{"key":"e_1_2_12_24_1","doi-asserted-by":"publisher","DOI":"10.1162\/rest_a_00693"},{"key":"e_1_2_12_25_1","volume-title":"Big data for 21st century economic statistics","author":"Chen J.","year":"2019"},{"key":"e_1_2_12_26_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1467-6419.2007.00518.x"},{"key":"e_1_2_12_27_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2013.10.002"},{"key":"e_1_2_12_28_1","first-page":"22","volume-title":"Model averaging and model selection","author":"Claeskens G.","year":"2008"},{"key":"e_1_2_12_29_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijforecast.2019.12.007"},{"key":"e_1_2_12_30_1","doi-asserted-by":"crossref","unstructured":"Cook T. &Smalter Hall A.(2017).Macroeconomic indicator forecasting with deep neural networks.","DOI":"10.2139\/ssrn.3046657"},{"key":"e_1_2_12_31_1","unstructured":"Coulombe P. G.(2019).Time\u2010varying parameters: A machine learning approach."},{"key":"e_1_2_12_32_1","doi-asserted-by":"publisher","DOI":"10.1214\/20-EJS1758"},{"key":"e_1_2_12_33_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2008.08.011"},{"key":"e_1_2_12_34_1","doi-asserted-by":"publisher","DOI":"10.2307\/2527342"},{"key":"e_1_2_12_35_1","doi-asserted-by":"publisher","DOI":"10.2307\/1392185"},{"key":"e_1_2_12_36_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2017.08.011"},{"key":"e_1_2_12_37_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijforecast.2018.09.006"},{"key":"e_1_2_12_38_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijforecast.2017.02.003"},{"key":"e_1_2_12_39_1","article-title":"Does the quantile regression forest learn more information on chinese systemic risk?","author":"Duan Y.","year":"2020","journal-title":"Available at SSRN 3556400"},{"key":"e_1_2_12_40_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijforecast.2015.11.017"},{"issue":"1","key":"e_1_2_12_41_1","first-page":"101","article-title":"A selective overview of variable selection in high dimensional feature space","volume":"20","author":"Fan J.","year":"2010","journal-title":"Statistica Sinica"},{"key":"e_1_2_12_42_1","unstructured":"Fortin\u2010Gagnon O. Leroux M. Stevanovic D. &Surprenant S.(2020).A large Canadian database for macroeconomic analysis."},{"key":"e_1_2_12_43_1","doi-asserted-by":"publisher","DOI":"10.1002\/jae.1177"},{"key":"e_1_2_12_44_1","doi-asserted-by":"publisher","DOI":"10.1162\/REST_a_00483"},{"key":"e_1_2_12_45_1","doi-asserted-by":"crossref","unstructured":"Giannone D. Lenza M. &Primiceri G. E.(2018).Economic predictions with big data: The illusion of sparsity.","DOI":"10.2139\/ssrn.3166281"},{"key":"e_1_2_12_46_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmoneco.2017.09.004"},{"key":"e_1_2_12_47_1","article-title":"To bag is to prune","author":"Goulet Coulombe P.","year":"2020","journal-title":"arXiv preprint arXiv:2008.07063"},{"key":"e_1_2_12_48_1","article-title":"Macroeconomic data transformations matter","author":"Goulet Coulombe P.","year":"2020","journal-title":"arXiv preprint arXiv:2008.01714"},{"key":"e_1_2_12_49_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0264-9993(03)00017-8"},{"key":"e_1_2_12_50_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijforecast.2019.01.010"},{"issue":"0","key":"e_1_2_12_51_1","article-title":"Autoencoder asset pricing models","volume":"0","author":"Gu S.","year":"2020","journal-title":"Journal of Econometrics"},{"key":"e_1_2_12_52_1","doi-asserted-by":"publisher","DOI":"10.1093\/rfs\/hhaa009"},{"key":"e_1_2_12_53_1","doi-asserted-by":"publisher","DOI":"10.3982\/ECTA5771"},{"key":"e_1_2_12_54_1","doi-asserted-by":"publisher","DOI":"10.3982\/ECTA10581"},{"key":"e_1_2_12_55_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-0-387-84858-7"},{"key":"e_1_2_12_56_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2016.03.006"},{"key":"e_1_2_12_57_1","unstructured":"Joseph A.(2019).Parametric inference with universal function approximators."},{"key":"e_1_2_12_58_1","doi-asserted-by":"publisher","DOI":"10.1257\/aer.20131193"},{"key":"e_1_2_12_59_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijforecast.2016.02.012"},{"key":"e_1_2_12_60_1","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.1999.10473882"},{"key":"e_1_2_12_61_1","doi-asserted-by":"publisher","DOI":"10.1002\/jae.2725"},{"issue":"1","key":"e_1_2_12_62_1","article-title":"Artificial neural networks: An econometric perspective","volume":"13","author":"Kuan C. M.","year":"1994","journal-title":"Econometric Reviews"},{"key":"e_1_2_12_63_1","unstructured":"Kuznetsov V. &Mohri M.(2015).Learning theory and algorithms for forecasting non\u2010stationary time series. InAdvances in neural information processing systems pp.541\u2013549."},{"key":"e_1_2_12_64_1","doi-asserted-by":"publisher","DOI":"10.1016\/0304-4076(93)90122-L"},{"key":"e_1_2_12_65_1","doi-asserted-by":"publisher","DOI":"10.1214\/16-STS588"},{"key":"e_1_2_12_66_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijforecast.2014.03.016"},{"key":"e_1_2_12_67_1","doi-asserted-by":"crossref","unstructured":"Litterman R. B.(1979).Techniques of forecasting using vector autoregressions.","DOI":"10.21034\/wp.115"},{"key":"e_1_2_12_68_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.dss.2009.02.001"},{"key":"e_1_2_12_69_1","doi-asserted-by":"publisher","DOI":"10.1002\/for.1059"},{"key":"e_1_2_12_70_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2005.07.020"},{"key":"e_1_2_12_71_1","doi-asserted-by":"publisher","DOI":"10.1080\/07350015.2015.1086655"},{"key":"e_1_2_12_72_1","doi-asserted-by":"publisher","DOI":"10.1002\/for.974"},{"issue":"0","key":"e_1_2_12_73_1","first-page":"1","article-title":"Forecasting inflation in a data\u2010rich environment: The benefits of machine learning methods","volume":"0","author":"Medeiros M. C.","year":"2019","journal-title":"Journal of Business and Economic Statistics"},{"issue":"6","key":"e_1_2_12_74_1","article-title":"Quantile regression forests.","volume":"7","author":"Meinshausen N.","year":"2006","journal-title":"Journal of Machine Learning Research"},{"key":"e_1_2_12_75_1","first-page":"1","article-title":"Forecasting Australia's real house price index: A comparison of time series and machine learning methods","author":"Milunovich G.","year":"2020","journal-title":"Journal of Forecasting"},{"key":"e_1_2_12_76_1","first-page":"789","article-title":"Stability bounds for stationary \u03d5$$ \\phi $$\u2010mixing and \u03b2$$ \\beta $$\u2010mixing processes","volume":"11","author":"Mohri M.","year":"2010","journal-title":"Journal of Machine Learning Research"},{"key":"e_1_2_12_77_1","doi-asserted-by":"publisher","DOI":"10.1002\/(SICI)1099-131X(200004)19:3<201::AID-FOR753>3.0.CO;2-4"},{"key":"e_1_2_12_78_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.econlet.2004.09.003"},{"key":"e_1_2_12_79_1","doi-asserted-by":"publisher","DOI":"10.1111\/caje.12070"},{"key":"e_1_2_12_80_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2014.07.040"},{"key":"e_1_2_12_81_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2014.10.031"},{"key":"e_1_2_12_82_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2013.04.002"},{"key":"e_1_2_12_83_1","doi-asserted-by":"publisher","DOI":"10.2307\/1391822"},{"key":"e_1_2_12_84_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2006.03.010"},{"key":"e_1_2_12_85_1","doi-asserted-by":"publisher","DOI":"10.1155\/2016\/4907654"},{"key":"e_1_2_12_86_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jeconom.2018.07.008"},{"key":"e_1_2_12_87_1","doi-asserted-by":"publisher","DOI":"10.1002\/for.2296"},{"key":"e_1_2_12_88_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijforecast.2018.01.001"},{"key":"e_1_2_12_89_1","doi-asserted-by":"crossref","unstructured":"Smola A. J. Murata N. Sch\u00f6lkopf B. &M\u00fcller K.\u2010R.(1998).Asymptotically optimal choice of\u03f5$$ \\epsilon $$\u2010loss for support vector machines. InInternational conference on artificial neural networks Springer London pp.105\u2013110.","DOI":"10.1007\/978-1-4471-1599-1_11"},{"key":"e_1_2_12_90_1","doi-asserted-by":"publisher","DOI":"10.1023\/B:STCO.0000035301.49549.88"},{"key":"e_1_2_12_91_1","first-page":"1","volume-title":"Cointegration, causality and forecasting: A festschrift for Clive W.J. Granger","author":"Stock J. H.","year":"1999"},{"key":"e_1_2_12_92_1","doi-asserted-by":"publisher","DOI":"10.1198\/016214502388618960"},{"key":"e_1_2_12_93_1","doi-asserted-by":"publisher","DOI":"10.1198\/073500102317351921"},{"key":"e_1_2_12_94_1","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/9780262013635.003.0003"},{"key":"e_1_2_12_95_1","doi-asserted-by":"publisher","DOI":"10.1353\/eca.2012.0005"},{"key":"e_1_2_12_96_1","doi-asserted-by":"publisher","DOI":"10.1080\/07350015.2012.715956"},{"key":"e_1_2_12_97_1","doi-asserted-by":"publisher","DOI":"10.1162\/003465397557123"},{"key":"e_1_2_12_98_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0169-2070(00)00065-0"},{"key":"e_1_2_12_99_1","doi-asserted-by":"publisher","DOI":"10.1016\/S1574-0706(05)01008-6"},{"key":"e_1_2_12_100_1","doi-asserted-by":"publisher","DOI":"10.1162\/089976600300015006"},{"issue":"4","key":"e_1_2_12_101_1","first-page":"1","article-title":"Proposing enhanced feature engineering and a selection model for machine learning processes","volume":"8","author":"Uddin M. F.","year":"2018","journal-title":"Applied Sciences (Switzerland)"},{"key":"e_1_2_12_102_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2010.08.004"},{"key":"e_1_2_12_103_1","unstructured":"Yousuf K. &Ng S.(2019).Boosting high dimensional predictive regressions with time varying parameters."},{"key":"e_1_2_12_104_1","first-page":"129","article-title":"Multiple kernel support vector regression for economic forecasting","author":"Zhang X. R.","year":"2010","journal-title":"2010 International Conference on Management Science and Engineering, ICMSE 2010"},{"issue":"0","key":"e_1_2_12_105_1","first-page":"1","article-title":"Causal interpretations of Black\u2010Box models","volume":"0","author":"Zhao Q.","year":"2019","journal-title":"Journal of Business and Economic Statistics"},{"key":"e_1_2_12_106_1","doi-asserted-by":"publisher","DOI":"10.1198\/016214506000000735"}],"container-title":["Journal of Applied Econometrics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/jae.2910","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/jae.2910","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/jae.2910","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,9,24]],"date-time":"2024-09-24T15:35:05Z","timestamp":1727192105000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/jae.2910"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,6,23]]},"references-count":105,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2022,8]]}},"alternative-id":["10.1002\/jae.2910"],"URL":"https:\/\/doi.org\/10.1002\/jae.2910","archive":["Portico"],"relation":{},"ISSN":["0883-7252","1099-1255"],"issn-type":[{"value":"0883-7252","type":"print"},{"value":"1099-1255","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,6,23]]},"assertion":[{"value":"2021-06-11","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-02-06","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-06-23","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}