Are the DSGE models is one of the econometric methods?

Hi all researchers,
Are the DSGE models is one of the econometric methods?

So there is a difference between models and methods, right? DSGE, AS-AD, IS-LM, Solow, OLG, Taylor rule, etc are all models of the following generic form: y_t = f(b, x_t, e_t). That is, exogenous x_t and/or residuals/shocks (e_t) explains endogenous y_t (ignoring model specific complexities like lags, etc), and b contains the parameters of the model.

We can now estimate parameters b of any model using econometrics methods. OLS, GMM, VAR, VECM, Bayesian methods, etc are econometrics methods that can be used to estimate economic models.

For example, we can use econometric methods like OLS or GMM to estimate parameters in a Taylor rule, or parameters in a reduced-form AS-AD or IS-LM model.

Same way, we can use econometric methods like GMM or Bayesian methods to estimate the parameters of a DSGE model.

So DSGE is not an econometrics method, it is an economic model (like OLG, IS-LM, AS-AD models) which can be estimated with econometric methods.

1 Like

Indeed. Let me add one more thing: DSGE models like any economic model will impose restrictions on the parameter values based on economics. Econometric models will be completely silent on this.


Thank you so much @kofiemma and @jpfeifer,
I was inspired by the following phrase (Kydland and Prescott, 1982): “further advances are needed before formal econometric methods can be fruitfully applied to testing this theory of aggregate fluctuations”.

What that quote meant is: you could employ formal econometric tests to the early RBC models at hand, but they always rejected the very simple models (see also In Praise of Calibration - Econlib). So they argued, we need more sophisticated models to better capture that data.

1 Like

Very clear