Your browser is not supported

Our website does not support the browser you are using. For a better browsing experience update to a compatible browser like the latest browsers from Chrome, Firefox and Safari.

Limited Information Estimation and Evaluation of DSGE Models

Martin Fukac, Adrian Pagan

We advance the proposition that dynamic stochastic general equilibrium (DSGE) models should not only be estimated and evaluated with full information methods. These require that the complete system of equations be specified properly. Some limited information analysis, which focuses upon specific equations, is therefore likely to be a useful complement to full system analysis. Two major problems occur when implementing limited information methods. These are the presence of forward-looking expectations in the system as well as unobservable non-stationary variables. We present methods for dealing with both of these difficulties, and illustrate the interaction between full and limited information methods using a well known model.
Fukač, Martin and Adrian Pagan (2010). ‘Limited information estimation and evaluation of DSGE models’, Journal of Applied Econometrics, Wiley, Volume 25(1), Pages 55-70, DOI: