Nadler, Daniel and Schmidt, Anatoly B., Market Impact of Macroeconomic Announcements: Do Surprises Matter? (June 11, 2014). Available for download at SSRN: http://ssrn.com/abstract=2449796
“Current research of market impact caused by macroeconomic announcements is based on regressing asset returns on macroeconomic surprises, S(t) ~ A(t) – E(t), where A(t) and E(t) are actual and expected (consensus) values of macroeconomic indicators at time t. We found that regressing returns of SPDR ETF SPY on actual change, AC(t) ~ A(t) – A(t-1), and expected change, EC(t) ~ E(t) – A(t-1), yield statistically significant indicators similar to those for S(t). These indicators somewhat vary for the periods 2004-2008 and 2009-2013. The advantage of AC(t) is that it is not based on the subjective nature of consensus. As for EC(t), it does not have a look-ahead bias since E(t) are published prior to A(t). Hence CE(t) can be used for short-term forecasting. We show that with proper fitting, AC-, EC-, and S-based indexes can be used for modeling the SPY price dynamics.”