Project information
Using textual data in DSGE models

This research proposes a novel integration of textual data extracted from US newspaper articles into dynamic stochastic general equilibrium (DSGE) models. Numerous recent studies have suggested that text-based indicators contain additional information not present in commonly used economic series. Using a hybrid model developed by Boivin and Giannoni (2006), we aim to investigate to what extent text-based indicators can provide relevant information about the dynamics of the economy in an estimated state-of-the-art DSGE model.