A Bayesian Approach to Macroeconomic Scenario Generation

Abstract

A number of central banks have come to rely on Dynamic Stochastic General Equilibrium (DGSE) models to inform their economic outlook and to help formulate their policy strategies. Some central banks that have developed DSGE models are the Bank of Canada, Bank of England, European Central Bank, Central Bank of Chile, Central Reserve of Peru and the US Federal Reserve . The objective of this research is to use the DSGE formulation to not only cast economic scenarios but also to generate risk assessments around those scenarios. The methodological approach consists of drawing simulations that reflect the correlation structure of the variables. The contribution would be twofold: (a) establish a methodology for the rank-ordering of scenarios based on the severity and the simulated distributions along the generated paths, (b) incorporate and integrate the two sources of uncertainty: parameter risk (Bayesian method) and model risk (distribution of errors).

A number of central banks have come to rely on Dynamic Stochastic General Equilibrium (DGSE) models to inform their economic outlook and to help formulate their policy strategies. Some central banks that have developed DSGE models are the Bank of Canada, Bank of England, European Central Bank, Central Bank of Chile, Central Reserve of Peru and the US Federal Reserve . The objective of this research is to use the DSGE formulation to not only cast economic scenarios but also to generate risk assessments around those scenarios. The methodological approach consists of drawing simulations that reflect the correlation structure of the variables. The contribution would be twofold: (a) establish a methodology for the rank-ordering of scenarios based on the severity and the simulated distributions along the generated paths, (b) incorporate and integrate the two sources of uncertainty: parameter risk (Bayesian method) and model risk (distribution of errors).