Title: DSGE models with expectations correction and directional forecast accuracy
Authors: Mauro Costantini - University of L'aquila (Italy) [presenting]
Abstract: The aim is to investigate the forecasting performance of a small scale New Keynesian model based on expectations correction. The forecast evaluation of the model is conducted using a measure of directional accuracy and directional value. This is the novelty of the paper. A Monte Carlo study is performed under different scenarios using DSGE and non-structural models. An empirical application to UK quarterly data over the period 1986-2019 is also carried out. The main results show that the small scale DSGE with expectations correction is competitive against non-structural models, both in terms of directional accuracy and directional value.