One of the advantages that VARs help us deal with is the issue of endogenity among variables. In terms of providing statistically sound forecasts it seems odd when compared to more static models.

Why is endogeneity no longer an issue when using the VAR framework?

  • $\begingroup$ I cannot understand the actual question. Could you perhaps rephrase it a bit? $\endgroup$ – Richard Hardy Jan 26 '18 at 17:30
  • $\begingroup$ @RichardHardy better? $\endgroup$ – EconJohn Jan 26 '18 at 17:39
  • $\begingroup$ Sorry, it was the first paragraph I could not understand, especially the second sentence. The answer to your title question is, regressors (lagged variables) cannot be affected by the regressand (non-lagged) because the present cannot affect the past. (Endogeneity would result if the regressand could affect the regressors.) $\endgroup$ – Richard Hardy Jan 26 '18 at 18:23
  • $\begingroup$ @RichardHardy can you post a detailed answer? I'd appreciate it. The text of title of the question is really the concern. $\endgroup$ – EconJohn Jan 26 '18 at 19:24

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.