Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Blanchard Quah Decomposition


From   Jorge Eduardo Pérez Pérez <perez.jorge@ur.edu.co>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: Blanchard Quah Decomposition
Date   Mon, 7 May 2012 17:46:09 -0400

After the SVAR estimation, type

irf set ir
irf create ir, step(40) set(ir) bs reps(1000)
irf cgraph (ir gdpadjust gdpadjust coirf) (ir gdpadjust uradjust oirf)
(ir uradjust gdpadjust coirf)   (ir uradjust uradjust oirf)

Notice that for the responses of output I draw the cumulative
responses and for unemployment the standard responses. This is because
output is included in differences in the VAR. Also notice that the
confidence intervals are different from the ones in the paper, apart
from being a different bootstrap, this happens because Stata uses a
95% C.I., while the paper uses 1 standard deviation (around 70%). You
can get difference confidence intervals using the -level- option.
_______________________
Jorge Eduardo Pérez Pérez


On Mon, May 7, 2012 at 2:32 PM, James Windsor-Clive
<jwindsorclive@googlemail.com> wrote:
> Thank you very much for that code, it should very useful. I am also
> looking to replicate figures (3)-(6) in the Blanchard and Quah paper.
> Do you have the code to create those figures?
>
> Kind regards
> James
>
> On Mon, May 7, 2012 at 7:04 PM, Jorge Eduardo Pérez Pérez
> <perez.jorge@ur.edu.co> wrote:
>> This code replicates Figure 8 of the Blanchard-Quah paper. I don't
>> know if there is an easier way to produce the structural shocks, I
>> created them manually. To use this code, you need the paper's dataset,
>> which is available here:
>>
>> http://www.estima.com/forum/viewtopic.php?f=4&t=304
>>
>> -----------------------
>> use "${dir}\datos\bqdata.dta"
>> tsset date
>>
>> gen loggnp=log(gnp)
>> gen loggd=log(gd87)
>> gen logrgnp=loggnp-loggd+log(100)
>> gen dlogrgnp=100*d.logrgnp
>>
>> * Extract separate means from the GNP growth series. Save the fitted
>> values for* rebuilding the data later.
>>
>> gen d1=(date<=tq(1973q4))
>> gen d2=(date>tq(1973q4))
>>
>> reg dlogrgnp d1 d2, nocons
>> predict gdpadjust, resid
>> predict means_from_gnp
>>
>> * gen gdpadjust=dlogrgnp
>> * Remove a linear trend from unemployment
>> gen trend=_n
>> reg lhmur trend
>> predict uradjust, resid
>> predict trend_from_ur
>>
>> matrix c=(.,0\.,.)
>> svar gdpadjust uradjust, lags(1/8) lreq(c)
>>
>> * Get structural shocks
>> matrix B=e(B)
>> predict e1, res eq(gdpadjust)
>> predict e2, res eq(uradjust)
>> mkmat e1 e2, matrix(e)
>> matrix eta=(inv(B)*e')'
>> svmat eta
>> * Zero out demand shocks and generate reduced form shocks again
>> replace eta2=0
>> mkmat eta1 eta2, matrix(eta)
>> matrix e=B*eta'
>> * Generate forecast with zeroed out shocks
>> matrix A1=e(A1)
>> matrix F=(inv(A1)*e)'
>> svmat F
>> * Add means back
>> replace F1=F1+means_from_gnp
>> replace F1=sum(F1)*0.01
>> sum date if e(sample)
>> local a=r(min)-1
>> * Add initial values back
>> sum logrgnp if date==`a'
>> replace F1=F1+r(mean)
>> ren F1 bq_trend
>> gen bq_gap=logrgnp-bq_trend
>> * Figure 8
>> tsline bq_gap if date>=tq(1950q1)
>> ---------------------------
>>
>> _______________________
>> Jorge Eduardo Pérez Pérez
>>
>>
>> On Mon, May 7, 2012 at 1:11 PM, James Windsor-Clive
>> <jwindsorclive@googlemail.com> wrote:
>>> Dear Statalist users,
>>>
>>> Apologies if you have answered this already but I am a relative
>>> newcomer to Stata and this list.
>>>
>>> I have conducted an SVAR using long run restrictions to decompose
>>> temporary and permanent shocks. I am using the methodology of
>>> Blanchard and Quah:
>>>
>>> Olivier Jean Blanchard; Danny Quah
>>> "The Dynamic Effects of Aggregate Demand and Supply Disturbances"
>>> The American Economic Review, Vol. 79, No. 4. (Sep., 1989), pp. 655-673
>>>
>>> I was wondering if there was a procedure on Stata that extracts the
>>> structural shock series from the estimated residual series. My aims is
>>> to manufacture impulse response functions showing the response of
>>> output to demand and supply shocks. Any guidance would be very much
>>> appreciated.
>>>
>>> Thanks in advance,
>>> James
>>> *
>>> *   For searches and help try:
>>> *   http://www.stata.com/help.cgi?search
>>> *   http://www.stata.com/support/statalist/faq
>>> *   http://www.ats.ucla.edu/stat/stata/
>>>
>>>
>>
>>
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/statalist/faq
>> *   http://www.ats.ucla.edu/stat/stata/
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>
>


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index