MDH, which attributes the time dependence of stock return volatility to that of
information flows. The lagged trading volume also plays a role in affecting conditional
volatility on both markets, with stronger results for the SPI 200 Futures.
In the final analysis, we investigate the news arrivals–volatility relation based on two
multivariate GARCH models: the Diagonal VECH and the Diagonal BEKK models.
The results of this analysis are presented in Table 12.6. From Table 12.6, we still observe
a positive relation between news arrivals and conditional volatility for both the S&P/
ASX 200 Index and the SPI 200 Futures and for the whole sample period as well as for
the subsample periods. News arrivals also increase the covariance between index and
futures returns. The results of the Wald Test reject the null hypothesis of insignificant
impacts of news arrivals on conditional volatility and covariance in both the Diagonal
VECH and Diagonal BEKK models. Thus, similar to Tables 12.3, 12.4, and 12.5, we
conclude that news arrivals have a positive impact on conditional volatility.
12.5 Conclusions
In the current study, we examine the effect of the rate of information arrivals on return
volatility, where the rate of information arrivals is proxied by the number of firm-
specific announcements per given interval. Investigating the S&P/ASX 200 Index and
the SPI 200 Futures over the period from October 2003 to September 2009, we docu-
ment that the rate of information arrivals has a positive impact on volatility. This
finding is consistent with the MDH, which argues that the variance of returns at a given
interval is proportional to the rate of information arrivals on the market. Moreover, the
level of volatility persistence is significantly reduced in both equity and futures markets
after controlling for the effect of news arrivals on volatility. Thus, similar to Lamoureux
and Lastrapes (1990), we argue that the phenomenon of volatility clustering reflects the
serial correlation of information arrival frequencies. Our empirical results are robust
with the use of univariate and multivariate conditional volatility modeling and during
the period before and after the global Credit Crisis. Future research could examine the
impact of different types (i.e., scheduled or unscheduled news announcements) or cat-
egories (i.e., mergers and acquisitions, earnings announcements, dividend announce-
ments, etc.) of news on volatility. The proposed investigation is motivated by the
observation of Andersen (1996) that different types of news may possess different arrival
processes, which in turn may generate different levels of short-term volatility dynamics.
12.A Technical appendix
We first test the relation between news arrivals and volatility of S&P/ASX 200 Index
returns and SPI 200 Futures returns by estimating the following censored regression
model:
jrtj¼ þ 1 Ntþ 2 Vt 1 þ"t; ð 12 : 2 Þin whichjrtjis either the absolute value of the seasonally adjusted return of the S&P/
ASX 200 Index or the SPI 200 Futures at thetth interval,Ntis the number of all
company announcements, andVt 1 is the lagged de-trended trading volume. The lagged
de-trended trading volume is also included to account for the potential that this variable
Firm-specific news arrival and the volatility of intraday stock index and futures returns 283