How does News Sentiment Impact Asset Volatility

Abstract:

This paper examines the dynamic relationship between firm-level return volatility and public news sentiment. By using the new RavenPack News Analytics - Dow Jones Edition database that captures over 1200 types of firm-specific and macroeconomic news releases and their sentiment scores at high frequencies, the author investigates the circumstances in which public news sentiment is related to the intraday volatility of the constituent stocks in the Dow Jones Composite Average (DJN 65).

Two different conditionally heteroskedastic models are employed: the Fractionally Integrated Generalized Autoregressive Conditionally Heteroskedastic (FIGARCH) and the twostate Markov Regime-Switching GARCH (RS-GARCH) models. For most of the DJN 65 stocks, our results confirm the significant impact of firm-specific news sentiment on intraday volatility persistence, even after controlling for the potential effects of macroeconomic news. Compared with macroeconomic news sentiment, firm-specific news sentiment apparently accounts for a greater proportion of overall volatility persistence.

Moreover, negative news has a greater impact on volatility than positive news. Furthermore, the results from the RS-GARCH model indicate that news sentiment accounts for a greater proportion of volatility persistence in the high-volatility regime (turbulent state) than in the low-volatility regime (calm state). In-sample forecasting performance and residual diagnostic tests suggest that FIGARCH generally outperforms RS-GARCH.

Introduction

The last few decades of research on finance have produced a tremendous number of papers focusing on the relationship between news flows and asset volatility. While some of these papers study the impact of firm-specific news releases on asset volatility (see, for example, Mitchell and Mulherin, 1994; Berry and Howe 1994; Kalev et al., 2004; and Grob-Klubmann and Hautsch, 2011), others explore the idea that assets react to macroeconomic news (Ederington and Lee, 1993; Andersen et al., 2003; Brenner et al., 2009; and Hautsch et al., 2011).

In part, these papers are theoretically motivated by the “Mixture of Distributions Hypothesis” (MDH), which provides an explanation for the relationship between the rate of information arrival and measures of market activity, such as asset volatility and trading volume. One implication of the MDH is that observed patterns of market activity are reflective of similar patterns in information flow (Clark, 1973; Epps and Epps, 1973; and Tauchen and Pitts, 1983). Furthermore, asset pricing models suggest that changes in risk factors (such as volatility) should affect asset returns, with the dynamic nature of these changes dictated by the dynamics ofnew information (such as public news releases) arriving in the market (Brenner et al., 2009).

Empirical evidence on the significance of firm-specific and macroeconomic news on asset volatility is mixed, however. For instance, Mitchell and Mulherin (1994) observe that the relation between volatility and information flows is statistically weak. Other researchers Kalev et al., 2004; Brenner et al., 2009; and Hautsch et al., 2011) find stronger evidence on the relation between firm-specific and macroeconomic announcements and asset volatility.

More recently, research has concentrated on the relationship between news sentiment and changes in asset dynamics (Tetlock, 2007, 2010; Tetlock et al., 2008; and Riordan et al., 2012). In particular, Riordan et al., (2012) argue that, compared with positive messages, the negative newswire messages are particularly informative and have a more significant impact on high-frequency asset price discovery and liquidity. However, these studies have not examined in detail how and when does news sentiment influence highfrequency asset volatility.

Preliminary evidence apparently suggests that the sentiment of firm-specific and macroeconomic news messages could be linked to high-frequency asset volatility . Figure 1 plots the hourly absolute returns (a proxy for asset return volatility) of four stocks from different industries: Bank of America (BAC), Coca-Cola (KO), General Electric (GE), and Microsoft (MSFT).

The shaded areas represent the period of arrival of intraday newswire messages, and the positive (negative) sentiment newswire messages are indicated in green (red). For KO and MSFT, multiple firm-specific and macroeconomic news releases occurring within a certain trading hour are apparently related to higher absolute returns during the same period. In the case of BAC, even though the number of news releases is smaller than KO and MSFT, a negative news release on credit ratings downgrade seems to be associated with substantially larger changes in the absolute returns These preliminary observations prompt us to address the following set of research questions.

First, what is the impact of intraday firm-specific and macroeconomic announcements on asset volatility in the proximity of their first release? To what extent can the sentiment of the high-frequency announcements account for persistence in asset volatility? Are there any differences between the impact of negative news sentiment and positive news sentiment on volatility? Is the impact of firm-specific announcements stronger or weaker than that of macroeconomic news? Put differently, after controlling for macroeconomic news sentiment, is firm-specific news sentiment still a significant explanatory variable for intraday asset volatility?

In addition, he also examines under what circumstances is news sentiment important for asset volatility. Specifically, by assuming that there are two possible states or regimes for volatility, he quantifies the impact of news sentiment in each of the states.