CMStatistics 2021: Start Registration
View Submission - CFE
Title: Impulse response functions in a self-exciting world Authors:  Daniel Soques - University of North Carolina Wilmington (United States) [presenting]
Neville Francis - University of North Carolina Chapel Hill (United States)
Michael Owyang - Federal Reserve Bank of St Louis (United States)
Abstract: Impulse response functions are calculated in situations where the data generating process depends upon a regime process that is driven by model variables. Researchers commonly use either generalized impulse response functions or local projections to estimate the true impulse response in such self-exciting models. A tradeoff exists between these two methods: local projections depend on the observed switches in the data while generalized impulse responses rely on specifying the correct regime process. We investigate under which conditions (i.e., model misspecification) each method is preferred. We then revisit the estimation of the fiscal multiplier using each method.