Miroljub Labus i Milica Labus su poslali članak "Decomposition of Impulse Response Functions in DSGE Models" na objavljivanje. U nastavku dajemo rezime na engleskom jeziku.
Impulse response functions (IRFs) are a standard tool for evaluating the performance of a Dynamic Stochastic General Equilibrium (DSGE) model. In complex DSGE models, however, IRFs may have unexpected shapes and paths, which cast doubts about the adjustment mechanisms of the underlying economy. We propose a method of decomposing each IRF into partial IRFs so that the individual contribution of state variables is clearly identified. By focusing on the state variables, it is straightforward to follow propagation of stochastic shocks in a DSGE model. We use DYNARE to solve DSGE models and provide a MATLAB file for IRFs decomposition. The entire process is illustrated by applying it to the canonic form of a Real Business Cycle (RBC) model. The same method can be used in any DSGE model of arbitrary complexity.