Hello
I have doubts about the transformation of some temporary variables in a data panel. I am trying to replicate some results of a paper that uses the local projections of Jordà (2005) combined with IV to generate fiscal multipliers of military spending (https://www.sciencedirect.com/scienc...47272722000330).
My doubt is related to how to generate the code according to the description in the text itself: "We normalize output and government spending by trend GDP. We estimate the trend as a quadratic time polynomial."
I also have doubts with the code to generate one of the regressions (second equation in the paper I attach), in which the authors estimate the fiscal multipliers for subsamples (developed vs. underdeveloped countries, for example), and test the statistical differences between the multipliers. They do this by interacting the regression coefficients with a subsample dummy.
Thank you very much for your help
I have doubts about the transformation of some temporary variables in a data panel. I am trying to replicate some results of a paper that uses the local projections of Jordà (2005) combined with IV to generate fiscal multipliers of military spending (https://www.sciencedirect.com/scienc...47272722000330).
My doubt is related to how to generate the code according to the description in the text itself: "We normalize output and government spending by trend GDP. We estimate the trend as a quadratic time polynomial."
I also have doubts with the code to generate one of the regressions (second equation in the paper I attach), in which the authors estimate the fiscal multipliers for subsamples (developed vs. underdeveloped countries, for example), and test the statistical differences between the multipliers. They do this by interacting the regression coefficients with a subsample dummy.
Thank you very much for your help
