News aware volatility forecasting: is the content of news important?
Robertson, Calum S., Geva, Shlomo, & Wolff, Rodney C. (2007) News aware volatility forecasting: is the content of news important? In Christen, Peter, Kennedy, Paul, Li, Jiuyong, Kolyshkina, Inna, & Williams, Graham (Eds.) Data mining and analytics 2007 : proceedings of the sixth Australasian Data Mining Conference (AusDM2007), Australian Computer Society in association with the ACM Digital LIbrary, Gold Coast, Queensland, pp. 157-166.
The efficient market hypothesis states that the market incorporates all available information to provide an accurate valuation of the asset at any given time. However, most models for forecasting the return or volatility of assets completely disregard the arrival of asset specific news (i.e., news which is directly relevant to the asset). In this paper we propose a simple adaptation to the GARCH model to make the model aware of news. We propose that the content of news is important and therefore describe a methodology to classify asset specific news based on the content. We present evidence from the US, UK and Australian markets which show that this model improves high frequency volatility forecasts. This is most evident for news which has been classified based on the content. We conclude that it is not enough to know when news is released, it is necessary to interpret its content.
Impact and interest:
Citation counts are sourced monthly from and citation databases.
These databases contain citations from different subsets of available publications and different time periods and thus the citation count from each is usually different. Some works are not in either database and no count is displayed. Scopus includes citations from articles published in 1996 onwards, and Web of Science® generally from 1980 onwards.
Citations counts from theindexing service can be viewed at the linked Google Scholar™ search.
Full-text downloads displays the total number of times this work’s files (e.g., a PDF) have been downloaded from QUT ePrints as well as the number of downloads in the previous 365 days. The count includes downloads for all files if a work has more than one.
Repository Staff Only: item control page