Setting shock magnitude in Impulse Response Function

[Luka Khimshiashvili]

Dears,

I have an issue with setting a shock magnitude size in Impulse Response Function (IRF). As I understand the Stata provides results for one standard deviation shock. Is it possible to convert standard deviation into percentage points? So that to have an answer how much percentage points is impulse changed when 1% change of impulse happens.

Thank you very much in advance.

[Jared Greathouse]

I don't understand the problem.

[Luka Khimshiashvili]

Dear Jared, Thank you for your reply.

Will try to clarify my problem.

I am using VAR estimation, and impulse response function (IRF) to get results for monetary shock's influence on inflation and GDPgrowth.

This is the result of my IRF. The units on the Y-axis is the standard deviation. So it shows what happens when the shock magnitude is one standard deviation. Can convert the units into percentage points, to get the effect of a 1pp interest rate shock on inflation/GDPgrowth in pp.

Thank you

[Jared Greathouse]

I still don't understand. Please, watch this. Or read this. Or both.

Using Statalist is just like making wild mushroom risotto. If I asked you to make my favorite version of risotto, you'd likely need to know what the SPECIFIC ingredients are and the steps you do to make said dish. So far you've given neither, you're asking us for help to make a dish, and you've not given the recipe (the code) or ingredients (the data). Both are required. You've just given figures and a sort of general description, which, while well intentioned, doesn't help anyone solve the problem. I can't even figure out which commands or syntax you used to produce the figures, and I couldn't reproduce them at all because you've not provided any example data to work with.
Precisely, you must give your dataset using dataex and your code that you've tried. No dataset, no code=no real assistance.

[Luka Khimshiashvili]

Jared, thank you again for reply.

Here is code of my estimation

tsset date1, q
dfuller Inflation
dfuller GDPrealgrowth
dfuller InterestRate
pperron Inflation
pperron GDPrealgrowth
pperron InterestRate
varsoc GDPrealgrowth Inflation InterestRate
var GDPrealgrowth Inflation InterestRate, lags(1/1)
varstable
predict error, resid
varlmar
vargranger
irf set irf9
irf create irf9, step(16)
irf graph irf, irf(irf9) impulse(GDPrealgrowth Inflation InterestRate) response(GDPrealgrowth Inflation InterestRate)

Here is my data

* Example generated by -dataex-. For more info, type help dataex
clear
input double(Inflation GDPrealgrowth InterestRate) int date1
 .027 -.049 .07456 196
 .021 -.087  .0608 197
-.008 -.015    .06 198
  .03     0   .056 199
 .047  .028    .05 200
 .044  .077  .0521 201
 .088  .069 .06533 202
 .105  .071  .0736 203
 .133  .074 .07744 204
 .126   .08    .08 205
 .067  .077 .07679 206
 .021  .065  .0718 207
-.013  .094 .06547 208
-.019  .073  .0618 209
    0  .054  .0575 210
-.006  .044   .056 211
-.019  .038  .0498 212
-.005  .033  .0432 213
-.006  .031 .03871 214
 .011  .043  .0375 215
 .033  .068 .03882 216
 .026  .048    .04 217
 .037  .044    .04 218
 .027  .025    .04 219
 .017  .043  .0427 220
 .035  .041   .048 221
 .051  .027 .05846 222
 .056  .014   .074 223
 .051  .034    .08 224
 .021  .045  .0755 225
 .009  .018 .06755 226
 .006  .022   .065 227
 .049  .068 .06684 228
 .066  .045  .0692 229
  .06  .044    .07 230
 .067  .041  .0705 231
 .033  .072  .0725 232
 .024   .06  .0725 233
 .029  .036 .07063 234
 .019  .032    .07 235
 .027  .052 .06775 236
 .044  .048   .065 237
 .053  .054 .06681 238
  .07  .046  .0837 239
end
format %tq date1

The last step in my estimation is generating graph of IRF, where the shock of magnitude equals 1 standard deviation as I understand. Please tell me if I am wrong. I want to have a magnitude of 1 percentage point change.

Thanks again.

[Luka Khimshiashvili]

UP