A1168
Title: Forecast relative error decomposition
Authors: Quinlan Lee - University of Toronto (Canada) [presenting]
Christian Gourieroux - University of Toronto and CREST (Canada)
Abstract: A class of relative error decomposition measures is introduced that are well-suited for the analysis of shocks in nonlinear dynamic models. They include the forecast relative error decomposition (FRED), forecast error Kullback decomposition (FEKD) and forecast error Laplace decomposition (FELD). These measures are favorable over the traditional forecast error variance decomposition (FEVD) because they account for nonlinear dependence in both a serial and cross-sectional sense. This is illustrated by applications to dynamic models for qualitative data, count data, stochastic volatility and cyber risk.
