My basic understanding of a DSGE is that it comuptes the equilibrium values for GDP, inflation, etc and then shows you how these would respond to different hypothetical exogeneous shocks.

How do you actually use this to forecast GDP and inflation etc over the next, say, 12 quarters?

Do you come up with a probability distribution of the possible exogeneous shocks (from past data), then put these into your DSGE model to produce a range of possible outcomes? (then compute the mean and standard deviation of GDP and inflation, from your simulations?)

  • $\begingroup$ In principle you could produce forecasts using DSGE models this way. As far as I'm aware it is mostly Bayesian methods being used for this purpose. The model is then adapted such that it is identified in econometric terms (by including enough shocks or a measurement error) and time series along with priors are used to estimate the model. Have a look at Herbst and Schorfheide (2015): Bayesian Estimation of DSGE models or Fernandez-Villaverde (2016): Solution and Estimation Methods for DSGE Models $\endgroup$
    – Joe
    Mar 24 at 16:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.