“…More generally, it is found that integrating survey-based forecasts within an estimated DSGE model can discipline model expectations and significantly mitigate the FG puzzle, while keeping the assumption of rational expectations. Müller, Christoffel, Mazelis and Montes-Galdón (2020) show that estimating a work-horse large policy model of the euro area, such as the new area wide model (NAWM) augmented with survey and financial market-based forecasts helps to contain the FG puzzle. It also improves the out-of-sample forecast accuracy of the resulting model (dubbed the "new area wide model with expectations", or NAWM-X).…”