“…where is the uncertainty component (i.e., aggregate uncertainty, common uncertainty or disagreement) of country i at month t, is a yearly dummy variable to capture time varying uncertainty, is a monthly dummy variable to capture seasonality in forecasts, is an inflation targeting dummy variable that takes a value of 1 during inflation targeting periods and zero otherwise, is a forward guidance dummy variable that takes a value of 1 during the forward guidance periods and zero otherwise, it M is a set of economic control variables, is the error term, , , , and are parameters, and is a coefficient vector associated with control variables. We include a number of control variables that have been identified in the literature as having potential impact on forecast uncertainty and disagreement (e.g., Dopke and Fritsche, 2006;van der Cruijsen and Demertzis, 2007;Patton and Timmermann, 2010;Dovern et al, 2012;Ehrmann et al, 2012;Lamla and Maag, 2012;Hartmann and Roestel, 2013;Posso and Tawadros, 2013;Siklos, 2013). Accordingly, our control variables are the lagged 1-month changes in dollar exchange rates, the change in WTI crude oil price, and unemployment rate, in addition to the output gap, the term spread defined as the difference between a 10-year government bond yield and a 3-month money market rate and the corresponding level variable for inflation, short-term or long-term rate.…”