I have some issues while discussing and interpreting this impulse response function (the graphics analysis). What do they mean and represent economically? What can the conclusions be?

Basically initially I had the time series of government spending (first column), GDP growth rate (second column) and inflation rate (third one).

Can you please help me understanding it? enter image description here


1 Answer 1


I would strongly encourage you to read a an introduction in VAR estimation, e.g. Lütkepohl. The visual impulse response analysis is quite simple: The columns always indicate the reaction to one shock. The first column gives the reaction to an one time expansive fiscal policy (GS-Shock). We see that GS increases up to 45 in period 0 and then decreases slowly in the following periods. As a reaction to this shock GDP and Inflation increase, the IRF is above 0 in the first periods. That is in line with the economic literature. The blue areas desrcibe the confidence intervals of your estimates. If the x axis is not included in your confidence interval you could conclude that your IRF is significantly different from 0.

Again please have a look in the literature. -> When you use the cholesky decomposition the ordering of your variables matter. Your ordering differs from previous literature. Normally one would assume that inflation and GDP react only with some delay.

  • $\begingroup$ Thanks , I’ll buy the book. But still for me is not clear the economic consequence of the third shock. Can you give me a hint? $\endgroup$ Commented Jan 30, 2019 at 11:47
  • $\begingroup$ The third column descibes an inflation shock. When inflation increases we see a small drop in GS in the third period. All other reactions are not signaificant different from zero. But the ordering might be "wrong". link(In literature GDP and inflation are more exogenous than goverment spending) $\endgroup$
    – Martin
    Commented Jan 30, 2019 at 13:14
  • $\begingroup$ I don’t know.. I just did it in the way the professor showed, but maybe as you say the ordering is wrong.. The topic/question was this one: Estimate a VAR model by OLS methods using U.S. quarterly data on government spending, GDP growth rate, inflation rate. Use a recursive identification scheme, i.e. Cholesky factorization scheme, to identify the structural shocks of the model. But thank you very much, I’ll try to manage the second shock now that I’ve understood le logic behind it. Thank you again! $\endgroup$ Commented Jan 31, 2019 at 14:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.