Course description

Regime Switching in VAR and DSGE Models: Theory and Applications


Regime switches are ubiquitous in economic data and can no longer be ignored in policy discussions. In a world with sometimes abrupt changes in uncertainty, breakdowns of economic structures and policy shifs (e.g unconventional policies), agents have to take into account the possibility of such events reoccurring.

This course will provide you with the necessary tools to analyse regime switches in economics. 

Course content

I) Constant parameter BVAR (brief introduction)

  • Identification and normalization
  • Linear restrictions and Gibbs sampling
  • Classical/Bayesian
  • Priors
  • Forecasting (unconditional and conditional)
  • Zero and sign restrictions
  • Application/Computer session

II) Switching BVAR

  • Exogenous swtiching VAR
  • Endogeous swtiching VAR
  • Simulation: Artificial data, irfs, variance decomposition, shock decomposition  
  • Application/Computer session

III) DSGE introduction

  • Solution methods
  • Perturbation
  • Taylor projections
  • Simulation: Artificial data, forecasting, variance decomposition
  • Filtering (linear and non-linear)
  • Estimation
  • Application/Computer session using RISE

IV) Time varying parameters in DSGE models

  • Exogenous switching
  • Simulation
  • Time varying parameters in DSGE models
  • Endogenous switching
  • Application/Computer session

V) Occasionally binding constraints

  • Optimal Simple rules
  • Optimal policy
  • Application/Computer session


Learning outcome knowledge

Course participants will master a range of tools for solving and estimating Bayesian Vector Autoregressive (BVAR) and Dynamic Stochastic General Equilibrium (DSGE) models in which parameters (or more general regimes) change subject to a switching process that is potentially endogenous. The course will be based on the Rationality In Switching Environments (RISE) Toolbox. 

Exam organisation

  • Written assignment: 100%