1. The colour of finance words
- Author
-
García, Diego, Hu, Xiaowen, and Rohrer, Maximilian
- Subjects
Machine learning ,Data mining ,Algorithms ,Data warehousing/data mining ,Algorithm ,Banking, finance and accounting industries ,Business ,Economics - Abstract
Keywords Measuring sentiment; Machine learning; Earnings calls; 10-Ks; WSJ Abstract Our paper relies on stock price reactions to colour words, in order to provide new dictionaries of positive and negative words in a finance context. We extend the machine learning algorithm of Taddy (2013), adding a cross-validation layer to avoid over-fitting. In head-to-head comparisons, our dictionaries outperform the standard bag-of-words approach (Loughran and McDonald, 2011) when predicting stock price movements out-of-sample. By comparing their composition, word-by-word, our method refines and expands the sentiment dictionaries in the literature. The breadth of our dictionaries and their ability to disambiguate words using bigrams both help to colour finance discourse better. Author Affiliation: (a) University of Colorado Boulder, United States (b) Southern Methodist University, United States (c) Norwegian School of Economics, Norway * Corresponding author. Article History: Received 9 December 2021; Revised 15 November 2022; Accepted 17 November 2022 (footnote)[white star] Dimitris Papanikolaou was the editor for this paper. We thank Simona Abis (discussant), Will Cong, Tony Cookson, Gerard Hoberg (discussant), Byoung-Hyoun Hwang (discussant), Jim Martin, David Stolin (discussant), Chenhao Tan, and Brian Waters for comments on an early draft, as well as seminar participants at Indiana University, INSEAD, the CU Boulder CS-NLP lab, the CU Boulder Finance division, the FutFinInfo webinar, NHH Finance brown bag, the 2021 FMA conference, the 2021 SFS Cavalcade, The Third Toronto Fintech Conference, the 2020 European Finance Association meetings, and the Michigan State University Fall 2019 conference. This work utilized the RMACC Summit supercomputer, which is supported by the National Science Foundation (Awards ACI-1532235 and ACI-1532236), the University of Colorado Boulder, and Colorado State University. The Summit supercomputer is a joint effort of the University of Colorado Boulder and Colorado State University. Byline: Diego García [http://leeds-faculty.colorado.edu/garcia/] (*,a), Xiaowen Hu [https://www.smu.edu/cox/Our-People-and-Community/Faculty/Xiaowen-Hu] (b), Maximilian Rohrer [https://www.maxrohrer.com] (c)
- Published
- 2023
- Full Text
- View/download PDF