NBER 基于近200年新闻文本的经济情绪分析

NBER WORKING PAPER SERIES(ALMOST) 200 YEARS OF NEWS-BASED ECONOMIC SENTIMENTJules H. van BinsbergenSvetlana BryzgalovaMayukh MukhopadhyayVarun SharmaWorking Paper 32026http://www.nber.org/papers/w32026NATIONAL BUREAU OF ECONOMIC RESEARCH1050 Massachusetts AvenueCambridge, MA 02138January 2024Any errors or omissions are the responsibility of the authors. For helpful comments, discussions, and suggestions, we thank Pedro Bordalo (discussant), Will Cong, Jingyu He (discussant), Manish Jha (discussant), Zhengyang Jiang, Asger Lunde (discussant), Alan Moreira (discussant), Maximilian Schleritzko (discussant), and Laura Veldkamp. We also thank the seminar and conference participants at London Business School, The Chinese University of Hong Kong, University of Surrey, 2023 Adam Smith Workshop, Macro Finance Society 22nd Workshop, Fifth International Workshop in Financial Econometrics, Georgia State FinTech Conference, Eastern Finance Association 2023, the CFA Society, and the Jacobs Levy Equity Management Center. We, the authors, have no relevant conflicts of interest to report regarding this research. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.© 2024 by Jules H. van Binsbergen, Svetlana Bryzgalova, Mayukh Mukhopadhyay, and Varun Sharma. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source.(Almost) 200 Years of News-Based Economic SentimentJules H. van Binsbergen, Svetlana Bryzgalova, Mayukh Mukhopadhyay, and Varun SharmaNBER Working Paper No. 32026January 2024JEL No. E2,E3,E4,E40,E43,E44,G01,G1,G10,G14,G17,G18,G40ABSTRACTUsing text from 200 million pages of 13,000 US local newspapers and machine learning methods, we construct a 170-year-long measure of economic sentiment at the country and state levels, that expands existing measures in both the time series (by more than a century) and the cross-section. Our measure predicts GDP (both nationally and locally), consumption, and employment growth, even after controlling for commonly-used predictors, as well as monetary policy decisions. Our measure is distinct from the information in expert forecasts and leads its consensus value. Interestingly, news coverage has become increasingly negative across all states in the past half-century.Jules H. van BinsbergenThe Wharton SchoolUniversity of Pennsylvania3620 Locust WalkPhiladelphia, PA 19104and NBERjulesv@wharton.upenn.eduSvetlana BryzgalovaLondon Business SchoolRegent's ParkLondon NW1 4SAsbryzgalova@london.eduMayukh Mukhopadhyay Sussex PlaceRegent's ParkLondon Business School London NW1 4SAUnited Kingdom mmukhopadhyay@london.edu