A0990
Title: Nonlinear impulse response functions and local projections
Authors: Christian Gourieroux - University of Toronto and CREST (Canada) [presenting]
Quinlan Lee - University of Toronto (Canada)
Abstract: The goal is to extend the non-parametric estimation of impulse response functions (IRF) by means of local projections in the nonlinear dynamic framework. The existence of a nonlinear autoregressive representation for Markov processes is discussed, and it is explained how their IRFs are directly linked to the nonlinear local projection (NLP), as in the case for the linear setting. A fully non-parametric LP estimator is presented in the one-dimensional nonlinear framework, and its asymptotic properties are compared to those of IRFs implied by the autoregressive model. Extensions are also considered to the multivariate framework through the lens of semiparametric models.
