Textual Changes in 10-Ks and Stock Price Crash Risk: Evidence from Neural Network Embeddings

28 Pages Posted: 22 Feb 2024

See all articles by Yahya Yilmaz

Yahya Yilmaz

University of Munster; University of Munster

Milan Reichmann

Leipzig University

Date Written: May 25, 2023

Abstract

Previous research attributes stock price crash risk to managerial bad news hoarding. Contrary to this notion, we find evidence that stock price crash risk is determined by investor inattention to textual changes in corporate disclosures. Using a large sample of 10-K filings, we estimate neural network embeddings to quantify the degree of textual changes in successive 10-Ks. We find that changes in 10-Ks have a positive and economically meaningful impact on one-year-ahead stock price crash risks. Our results suggest that investor inattention to textual changes in 10-Ks can have broader capital market consequences than previously documented.

Suggested Citation

Yilmaz, Yahya and Reichmann, Milan, Textual Changes in 10-Ks and Stock Price Crash Risk: Evidence from Neural Network Embeddings (May 25, 2023). Proceedings of the EUROFIDAI-ESSEC Paris December Finance Meeting 2023, Available at SSRN: https://ssrn.com/abstract=4643560 or http://dx.doi.org/10.2139/ssrn.4643560

Yahya Yilmaz (Contact Author)

University of Munster ( email )

Germany

University of Munster ( email )

Germany

Milan Reichmann

Leipzig University ( email )

Leipzig, DE
Germany

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
44
Abstract Views
200
PlumX Metrics