Media type: E-Article Title: Volatility prediction based on scheduled macroeconomic announcements Contributor: Petralias, Athanassios; Dellaportas, Petros imprint: Wiley, 2015 Published in: Canadian Journal of Statistics Language: English DOI: 10.1002/cjs.11247 ISSN: 0319-5724; 1708-945X Keywords: Statistics, Probability and Uncertainty ; Statistics and Probability Origination: Footnote: Description: <jats:title>Abstract</jats:title><jats:sec><jats:label /><jats:p>We investigate the impact of scheduled macroeconomic announcements on the volatility of exchange rates by introducing a flexible model formulation. For each macroeconomic index we estimate cutoff points in the surprise component of the announcement that specify the degree the volatility process is affected. This degree is quantified by jumps of unknown size that occur before and at the time of the announcement and then die out exponentially with unknown rate. We make inferences using a population Markov chain Monte Carlo reversible jump algorithm and illustrate our methodology by predicting exchange rates volatility using 15 U.S. macroeconomic announcements. <jats:italic>The Canadian Journal of Statistics</jats:italic> 43: 199–223; 2015 © 2015 Statistical Society of Canada</jats:p></jats:sec>