Title: Mind the gap: Stylized dynamic facts and structural models
Authors: Filippo Ferroni - Federal Reserve Bank of Chicago (United States) [presenting]
Fabio Canova - Norwegian Business School (Norway)
Abstract: The purpose is to study what happens to identified shocks and to dynamic responses when the structural model features $q$ disturbances and $m$ endogenous variables, $q=m$, but only $m_1 < q$ variables are used in the empirical model. Aggregation create problems. Appropriate theoretical restrictions may be insufficient to obtain the structural disturbances and the dynamics they produce. Identified shocks do not necessarily combine structural disturbances of the same type. Instead, they are linear combinations of current and past values of all structural disturbances. The theory used to interpret the data and the disturbances it features determines whether an empirical model is too small or not. An example highlights the magnitude of the distortions and the steps needed to reduce them. We revisit previous evidence regarding the transmission of house price shocks.