News sentiment and states of stock return volatility

Evidence from long memory and discrete choice models

Yanlin Shi*, Kin Yip Ho

*Corresponding author for this work

Research output: Contribution to journalArticle


This paper examines the impact of public news sentiment on volatility states of firm-level returns. We firstly propose a Markov regime switching fractionally integrated exponential GARCH (MRS-FIEGARCH) model, which is employed to estimate the latent volatility states of intraday stock return. By using the new RavenPack Dow Jones News Analytics database, we fit discrete choice models to investigate the impact of news sentiment on changes of volatility states of the constituent stocks in the Dow Jones Composite Average. Our results demonstrate that sentiments of macroeconomic and firm-specific news can significantly influence the likelihoods of volatility states of intraday stock returns.

Original languageEnglish
Article number101446
JournalFinance Research Letters
Publication statusE-pub ahead of print - 1 Jan 2020



  • Asset volatility
  • Discrete choice model
  • Markov regime-switching FIEGARCH
  • News sentiment
  • Public information arrival

Cite this