{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T20:13:34Z","timestamp":1788293614489,"version":"build-2803163510"},"reference-count":48,"publisher":"The Econometric Society","issue":"4","license":[{"start":{"date-parts":[[2024,1,1]],"date-time":"2024-01-01T00:00:00Z","timestamp":1704067200000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/legalcode"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["QE"],"abstract":"<jats:p>We study the Latent Dirichlet Allocation model, a popular Bayesian algorithm for text analysis. We show that the model's parameters are not identified, which suggests that the choice of prior matters. We characterize the range of values that the posterior mean of a given functional of the model's parameters can attain in response to a change in the prior, and we suggest two algorithms that report this range. Both of our algorithms rely on obtaining multiple \n            <jats:italic>Nonnegative Matrix Factorizations<\/jats:italic> of either the posterior draws of the corpus' population term\u2010document frequency matrix or of its maximum likelihood estimator. The key idea is to maximize\/minimize the functional of interest over all these nonnegative matrix factorizations. To illustrate the applicability of our results, we revisit recent work studying the effects of increased transparency on the communication structure of monetary policy discussions in the United States.\n          <\/jats:p>","DOI":"10.3982\/qe1825","type":"journal-article","created":{"date-parts":[[2024,11,15]],"date-time":"2024-11-15T09:48:52Z","timestamp":1731664132000},"page":"939-970","source":"Crossref","is-referenced-by-count":8,"title":["Robust machine learning algorithms for text analysis"],"prefix":"10.3982","volume":"15","author":[{"given":"Shikun","family":"Ke","sequence":"first","affiliation":[{"name":"Yale School of Management, Yale University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jos\u00e9 Luis","family":"Montiel Olea","sequence":"additional","affiliation":[{"name":"Department of Economics, Cornell University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"James","family":"Nesbit","sequence":"additional","affiliation":[{"name":"Distribution Optimization and Grocery Innovation, Amazon F3"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"2359","published-online":{"date-parts":[[2024]]},"reference":[{"key":"quan200329-bib-0001","doi-asserted-by":"crossref","first-page":"1582","DOI":"10.1137\/130913869","volume":"45","author":"Arora","year":"2016","journal-title":"SIAM Journal on Computing"},{"key":"quan200329-bib-0002","series-title":"FOCS'12","first-page":"1","volume-title":"Proceedings of the 2012 IEEE 53rd Annual Symposium on Foundations of Computer Science","author":"Arora","year":"2012"},{"key":"quan200329-bib-0003","doi-asserted-by":"crossref","first-page":"1593","DOI":"10.1093\/qje\/qjw024","volume":"131","author":"Baker","year":"2016","journal-title":"The Quarterly Journal of Economics"},{"key":"quan200329-bib-0004","doi-asserted-by":"crossref","first-page":"1325","DOI":"10.1086\/705331","volume":"128","author":"Bandiera","year":"2020","journal-title":"Journal of Political Economy"},{"key":"quan200329-bib-0005","doi-asserted-by":"crossref","first-page":"303","DOI":"10.1016\/0378-3758(90)90079-A","volume":"25","author":"Berger","year":"1990","journal-title":"Journal of Statistical Planning and Inference"},{"key":"quan200329-bib-0006","doi-asserted-by":"crossref","first-page":"2189","DOI":"10.3982\/ECTA16581","volume":"89","author":"Bhattacharya","year":"2021","journal-title":"Econometrica"},{"key":"quan200329-bib-0007","doi-asserted-by":"crossref","first-page":"1765","DOI":"10.3150\/19-BEJ1166","volume":"26","author":"Bing","year":"2020","journal-title":"Bernoulli"},{"key":"quan200329-bib-0008","volume":"21","author":"Bing","year":"2020","journal-title":"Journal of Machine Learning Research"},{"key":"quan200329-bib-0009","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1145\/2133806.2133826","volume":"55","author":"Blei","year":"2012","journal-title":"Communications of the ACM"},{"key":"quan200329-bib-0010","doi-asserted-by":"crossref","first-page":"859","DOI":"10.1080\/01621459.2017.1285773","volume":"112","author":"Blei","year":"2017","journal-title":"Journal of the American Statistical Association"},{"key":"quan200329-bib-0011","first-page":"17","volume":"1","author":"Blei","year":"2007","journal-title":"The Annals of Applied Statistics"},{"key":"quan200329-bib-0012","first-page":"34","volume-title":"Text Mining: Classification, Clustering, and Applications","volume":"10","author":"Blei","year":"2009"},{"key":"quan200329-bib-0013","first-page":"993","volume":"3","author":"Blei","year":"2003","journal-title":"Journal of Machine Learning Research"},{"key":"quan200329-bib-0014","doi-asserted-by":"crossref","first-page":"561","DOI":"10.1145\/2940716.2940787","author":"Budak","year":"2016","journal-title":"Proceedings of the 2016 ACM Conference on Economics and Computation"},{"key":"quan200329-bib-0015","volume-title":"Doeblin and Modern Probability","volume":"149","author":"Doeblin","year":"1993"},{"key":"quan200329-bib-0016","first-page":"1141","volume":"16","author":"Donoho","year":"2004","journal-title":"Advances in Neural Information Processing Systems"},{"key":"quan200329-bib-0017","volume-title":"Mathematical Statistics: A Decision Theoretic Approach","volume":"7","author":"Ferguson","year":"1967"},{"key":"quan200329-bib-0018","unstructured":"Freyaldenhoven, Simon, Shikun Ke, Dingyi Li, and Jos\u00e9 Luis Montiel Olea (2023), \u201cOn the testability of the anchor words assumption in topic models.\u201d Working Paper, Cornell University."},{"key":"quan200329-bib-0019","doi-asserted-by":"crossref","first-page":"2145","DOI":"10.1214\/aos\/1034713651","volume":"23","author":"Ghosal","year":"1995","journal-title":"The Annals of Statistics"},{"key":"quan200329-bib-0020","doi-asserted-by":"crossref","unstructured":"Giacomini, Raffaella, Toru Kitagawa, and Harald Uhlig (2019), \u201cEstimation under ambiguity.\u201d Technical Report, Cemmap Working Paper.","DOI":"10.1920\/wp.cem.2019.2419"},{"key":"quan200329-bib-0021","doi-asserted-by":"crossref","first-page":"1519","DOI":"10.3982\/ECTA16773","volume":"89","author":"Giacomini","year":"2021","journal-title":"Econometrica"},{"key":"quan200329-bib-0022","first-page":"1","volume":"19","author":"Giordano","year":"2018","journal-title":"Journal of Machine Learning Research"},{"key":"quan200329-bib-0023","doi-asserted-by":"crossref","first-page":"5228","DOI":"10.1073\/pnas.0307752101","volume":"101","author":"Griffiths","year":"2004","journal-title":"Proceedings of the National Academy of Sciences"},{"key":"quan200329-bib-0024","doi-asserted-by":"crossref","first-page":"1682","DOI":"10.1198\/jasa.2009.tm08603","volume":"104","author":"Gustafson","year":"2009","journal-title":"Journal of the American Statistical Association"},{"key":"quan200329-bib-0025","doi-asserted-by":"crossref","first-page":"801","DOI":"10.1093\/qje\/qjx045","volume":"133","author":"Hansen","year":"2018","journal-title":"The Quarterly Journal of Economics"},{"key":"quan200329-bib-0026","first-page":"856","volume":"23","author":"Hoffman","year":"2010","journal-title":"Advances in Neural Information Processing Systems"},{"key":"quan200329-bib-0027","doi-asserted-by":"crossref","unstructured":"Ke, Shikun, Jos\u00e9 Luisb Montiel Olea, and James Nesbit (2024), \u201cSupplement to \u2018Robust machine learning algorithms for text analysis\u2019.\u201d Quantitative Economics Supplemental Material, 15, https:\/\/doi.org\/10.3982\/QE1825.","DOI":"10.3982\/QE1825"},{"key":"quan200329-bib-0028","doi-asserted-by":"crossref","unstructured":"Ke, Zheng Tracy, Bryan T. Kelly, and Dacheng Xiu (2019), \u201cPredicting returns with text data.\u201d National Bureau of Economic Research Working paper w26186.","DOI":"10.3386\/w26186"},{"issue":"545","key":"quan200329-bib-0029","doi-asserted-by":"crossref","first-page":"434","DOI":"10.1080\/01621459.2022.2123813","volume":"119","author":"Ke","year":"2024","journal-title":"Journal of the American Statistical Association,"},{"key":"quan200329-bib-0030","doi-asserted-by":"crossref","first-page":"165","DOI":"10.1214\/aoms\/1177729837","volume":"21","author":"Koopmans","year":"1950","journal-title":"The Annals of Mathematical Statistics"},{"key":"quan200329-bib-0031","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1155\/2008\/764206","volume":"2008","author":"Laurberg","year":"2008","journal-title":"Computational Intelligence and Neuroscience"},{"key":"quan200329-bib-0032","doi-asserted-by":"crossref","first-page":"257","DOI":"10.1137\/20M1378971","volume":"43","author":"Laursen","year":"2022","journal-title":"SIAM Journal on Matrix Analysis and Applications"},{"key":"quan200329-bib-0033","first-page":"556","volume":"13","author":"Lee","year":"2001","journal-title":"Advances in Neural Information Processing Systems"},{"key":"quan200329-bib-0034","doi-asserted-by":"crossref","first-page":"695","DOI":"10.1111\/j.1468-0297.2008.02138.x","volume":"118","author":"Meade","year":"2008","journal-title":"The Economic Journal"},{"key":"quan200329-bib-0035","doi-asserted-by":"crossref","first-page":"755","DOI":"10.3982\/ECTA8360","volume":"80","author":"Moon","year":"2012","journal-title":"Econometrica"},{"key":"quan200329-bib-0036","doi-asserted-by":"crossref","first-page":"358","DOI":"10.1017\/S0003055417000570","volume":"112","author":"Mueller","year":"2018","journal-title":"American Political Science Review"},{"key":"quan200329-bib-0037","doi-asserted-by":"crossref","first-page":"256","DOI":"10.1080\/07350015.2020.1802285","volume":"40","author":"Munro","year":"2022","journal-title":"Journal of Business & Economic Statistics"},{"key":"quan200329-bib-0038","doi-asserted-by":"crossref","first-page":"716","DOI":"10.1016\/j.jeconom.2020.06.008","volume":"222","author":"Montiel Olea","year":"2021","journal-title":"Journal of Econometrics"},{"key":"quan200329-bib-0039","doi-asserted-by":"crossref","first-page":"111","DOI":"10.1002\/env.3170050203","volume":"5","author":"Paatero","year":"1994","journal-title":"Environmetrics"},{"key":"quan200329-bib-0040","doi-asserted-by":"crossref","first-page":"483","DOI":"10.1017\/S0266466698144043","volume":"14","author":"Poirier","year":"1998","journal-title":"Econometric Theory"},{"key":"quan200329-bib-0041","doi-asserted-by":"crossref","first-page":"577","DOI":"10.2307\/1913267","volume":"39","author":"Rothenberg","year":"1971","journal-title":"Econometrica"},{"key":"quan200329-bib-0042","doi-asserted-by":"crossref","first-page":"1566","DOI":"10.1198\/016214506000000302","volume":"101","author":"Teh","year":"2006","journal-title":"Journal of the American Statistical Association"},{"key":"quan200329-bib-0043","first-page":"1973","volume":"22","author":"Wallach","year":"2009","journal-title":"Advances in Neural Information Processing Systems"},{"key":"quan200329-bib-0044","doi-asserted-by":"crossref","first-page":"1387","DOI":"10.1214\/aos\/1176347277","volume":"17","author":"Wasserman","year":"1989","journal-title":"The Annals of Statistics"},{"key":"quan200329-bib-0045","first-page":"465","volume":"31","author":"Watson","year":"2016","journal-title":"Statistical Science"},{"key":"quan200329-bib-0046","first-page":"1151","volume-title":"Proceedings of the 27th International Conference on Machine Learning (ICML-10)","author":"Williamson","year":"2010"},{"key":"quan200329-bib-0047","first-page":"3455","volume":"27","author":"Zhou","year":"2014","journal-title":"Advances in Neural Information Processing Systems"},{"key":"quan200329-bib-0048","first-page":"3043","volume":"28","author":"Zhou","year":"2015","journal-title":"Advances in Neural Information Processing Systems"}],"container-title":["Quantitative Economics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.3982\/QE1825","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,11,15]],"date-time":"2024-11-15T09:49:23Z","timestamp":1731664163000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.econometricsociety.org\/doi\/10.3982\/QE1825"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024]]},"references-count":48,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2024]]}},"alternative-id":["10.3982\/QE1825"],"URL":"https:\/\/doi.org\/10.3982\/qe1825","relation":{},"ISSN":["1759-7323"],"issn-type":[{"value":"1759-7323","type":"print"}],"subject":[],"published":{"date-parts":[[2024]]}}}