Discretionary dissemination on Twitter
Using an unsupervised machine learning approach to analyze 12.8 million tweets posted by S&P 1500 firms from 2012 to 2016, we find that firms tweet more financial information around significantly negative or positive earnings announcements or accounting filings. Specifically, we observe a symmet...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | text |
| Language: | English |
| Published: |
Institutional Knowledge at Singapore Management University
2020
|
| Subjects: | |
| Online Access: | https://ink.library.smu.edu.sg/soa_research/1776 https://ink.library.smu.edu.sg/context/soa_research/article/2803/viewcontent/SSRN_id3105847.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Singapore Management University |
| Language: | English |
| Summary: | Using an unsupervised machine learning approach to analyze 12.8 million tweets posted by S&P 1500 firms from 2012 to 2016, we find that firms tweet more financial information around significantly negative or positive earnings announcements or accounting filings. Specifically, we observe a symmetric U-shaped relation between the number of financial tweets and the materiality of accounting information events. This relation is consistent with the theoretical prediction in Hummel et al. (2018) which assumes that managers are sensitive to their firm’s fundamental value. We document that this relation also holds for hyperlink usage in tweets about financial information around important events, and that the relation is more pronounced for non-loss firms. Furthermore, our intraday analyses indicate that firms release financial information on Twitter primarily after (before) earnings announcements (10-K or 10-Q filings), suggesting that Twitter plays different roles for firms around separate accounting information events. |
|---|