?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.relation=http%3A%2F%2Feprints.imtlucca.it%2F4082%2F&rft.title=Predicting+Exporters+with+Machine+Learning&rft.creator=Micocci%2C+Francesca&rft.creator=Rungi%2C+Armando&rft.subject=HB+Economic+Theory&rft.subject=HF+Commerce&rft.subject=HG+Finance&rft.description=In+this+contribution%2C+we+exploit+machine+learning+techniques+to+predict+out-of-sample+firms'%0D%0Aability+to+export+based+on+the+financial+accounts+of+both+exporters+and+non-exporters.+Therefore%2C%0D%0Awe+show+how+forecasts+can+be+used+as+exporting+scores%2C+i.e.%2C+to+measure+the+distance+of%0D%0Anon-exporters+from+export+status.+For+our+purpose%2C+we+train+and+test+various+algorithms+on+the%0D%0Afinancial+reports+of+57%2C021+manufacturing+firms+in+France+in+2010-2018.+We+find+that+a+Bayesian%0D%0AAdditive+Regression+Tree+with+Missingness+In+Attributes+(BART-MIA)+performs+better+than%0D%0Aother+techniques+with+a+prediction+accuracy+of+up+to+0%3A90.+Predictions+are+robust+to+changes+in%0D%0Adefinitions+of+exporters+and+in+the+presence+of+discontinuous+exporters.+Eventually%2C+we+argue%0D%0Athat+exporting+scores+can+be+helpful+for+trade+promotion%2C+trade+credit%2C+and+to+assess+firms'%0D%0Acompetitiveness.+For+example%2C+back-of-the-envelope+estimates+show+that+a+representative+firm%0D%0Awith+just+below-average+exporting+scores+needs+up+to+44%25+more+cash+resources+and+up+to+2%3A5%0D%0Atimes+more+capital+expenses+to+reach+full+export+status.&rft.date=2021-07&rft.type=Working+Paper&rft.type=NonPeerReviewed&rft.format=application%2Fpdf&rft.language=en&rft.rights=cc_by_nc&rft.identifier=http%3A%2F%2Feprints.imtlucca.it%2F4082%2F1%2FWP_EIC_3-2021.pdf&rft.identifier=++Micocci%2C+Francesca+and+Rungi%2C+Armando++Predicting+Exporters+with+Machine+Learning.++EIC+working+paper+series++%233%2F2021+++++ISSN+2279-6894.++++++