CMStatistics 2018: Start Registration
View Submission - CFE
Title: Multivariate quantile impulse response functions Authors:  Gabriel Montes-Rojas - Universidad de Buenos Aires (Argentina) [presenting]
Abstract: A multivariate vector autoregression quantile (VARQ) model is developed which is the solution to a collection of directional quantile models for a fixed orthonormal basis, in which each component represents a directional quantile that corresponds to a particular endogenous variable. This corresponds to a reduced form multivariate quantile autoregressive model is developed to study heterogeneity in the effects of macroeconomic shocks. The VARQ estimator allows us to forecast the future performance of the multivariate time-series conditional on the available information, which depends on multivariate quantile indexes. From the forecasting procedure we define an impulse-response function (IRF) model that computes the effect of a given perturbation in a (some) variable(s). This thus generalizes the mean-based IRF analysis to the multivariate quantile framework. This analysis explores potential dynamic heterogeneity not covered by the mean-based IRF analysis using mean-based VAR. In particular we can study the realization of particular sequences of events, as defined by particular values of the multivariate quantile indexes, defined as quantile paths. The model is applied to study monetary shocks in a three-variable macroeconomic model (output gap, inflation, Fed Funds rate) for the U.S. for the period 1980 to 2010.