I'm wondering what methods are used in economics, specially in macro, to deal with model misspecification.
I've already searched the web, but I may be missing some methods. Therefore, I ask you for some help.
Economics Stack Exchange is a question and answer site for those who study, teach, research and apply economics and econometrics. It only takes a minute to sign up.Sign up to join this community
Thomas Sargent and Lars Peter Hansen have a series of work on "robustness", a theory (and technique) that deals directly with potential model misspecifications in the context of macro. [The term "robustness" derives from robust control theory].
The standard theory of decision making under uncertainty advises the decision maker to form a statistical model linking outcomes to decisions and then to choose the optimal distribution of outcomes. This assumes that the decision maker trusts the model completely. But what should a decision maker do if the model cannot be trusted?
Lars Hansen and Thomas Sargent, two leading macroeconomists, push the field forward as they set about answering this question. They adapt robust control techniques and apply them to economics. By using this theory to let decision makers acknowledge misspecification in economic modeling, the authors develop applications to a variety of problems in dynamic macroeconomics.
Technical, rigorous, and self-contained, this book will be useful for macroeconomists who seek to improve the robustness of decision-making processes.
I think it really depends on what your views on model misspecification are. How do you determine whether your model is misspecified? Testing is not really feasible in a Bayesian framework which is increasingly common in macroeconomics nor is it really considered when you have stuctural or theory-based models. Oftentimes the closest people get to dealing with this is to use HAC standard errors to deal with heteroskedasticy and autocorrelation. In empirical VAR's the main approaches are lag selection to address autocorrelation concerns and either differencing or using a Minnesota type prior to address non-stationarity concerns (or going the cointegration route which also addresses non-stationarity). The David Hendry school of thought takes a different approach by incorporating testing for misspecification into the model selection / model design procedure and including non-linear transformations of the data, various indicator saturation procedures and lags.