I read several papers talking about the so called dust settling period for a impulse response function (IRF) derived from VAR.

e.g., Nijs, Vincent R., et al. "The category-demand effects of price promotions." Marketing science 20.1 (2001): 1-22.

Basically, the dust settling period refers to the time interval before convergence of IRF is obtained. The key is to find the point at which IRF becomes stable or reaches convergence level. But the papers do not provide detailed description about how to find such point. Can any one help or refer to some useful reading material? I googled but was unable to find useful stuff.

  • $\begingroup$ I would assume that you could just look at the graph and find the point where the IRF reaches approximately zero, while also being stable (not growing afterwards). $\endgroup$ – Dole Jan 3 '16 at 15:49
  • $\begingroup$ @Dole, Thanks for your comment. But I am wondering is there a more mathematically precise way to determine the steady point? Research paper requires more rigorous method to present the results rather than simply look at the graph. $\endgroup$ – Tracy Yang Jan 3 '16 at 17:09
  • $\begingroup$ mathematically, for difference equations convergence means the point where the function reaches a given interval and does not fluctuate from it. This often takes an infinite time, so you will have to use your best judgement (ie. when does the effect become economically irrelevant). You could bootstrap and use the confidence interval as your guide $\endgroup$ – Dole Jan 3 '16 at 18:29

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.