Re: [Gretl-users] impulse-response functions from VECM in Gretl
by Ioannis A. Venetis
=======================
Question
So what about my last question? Are IRF confidence intervals based on
bootstrapping valid even if the VECM has non-normal residuals?
Answer
The bootstrap was developed to avoid Monte Carlo simulations using
random draws from a preset distribution (say Normal). The bootstrap will
proxy the empirical distribution of your residuals given that the
residuals are "sufficiently" noisy (it doesn't matter if they test
non-normal!)
By noisy, I mean no heteroscedasticity or autocorrelation issues. There
is a huge literature for heteroscedasticity issues (Wild bootstrap) and
I guess (i am not an expert) block bootstrap that divides your data into
blocks to preserve the original time series properties within a block
(when dependence is an issue).
Back to your question, it seems it has to do with what you mean
"non-normal" residuals? My guess is a few outliers or asymmetry? So
bootstrap is valid.
=======================
Question d.o.f correction
Regarding d.o.f corrections when calculating IRFs in gretl
Using Allin's example
open data9-7
var 4 UNEMP PRIME
For the UNEMP equation the Sum squared resid=3.613493 so that
sqrt(3.613493/60)=0.24540759
Notice that this is the one standard error shock in UNEMP and 60 is the
original sample size of T=64 obs minus p=4 for the lags (say T-p).
In other commercial software this is baptized : no df correction.
If you prefer df correction multiply IRFs with 60/(60-1-8) where 1+8
corresponds to the parameters number in one VAR equation
sqrt(3.613493/(60-9))=0.2661819
Now I think the SVAR addon uses the correction
sqrt(3.613493/(64-1-8))=0.25631986 that is the original sample size T
minus the number of parameters in the UNEMP equation
==========================
6 years, 8 months
impulse-response functions from VECM in Grtel
by lasses skola
Hi! I would like to know what unit is on the left y-axis in the graphs
containing impulse-resonse functions, calculated using bootstrapping from
VEC models in Gretl. Can someone please help me with this?
Best regards,
Lars Ahnland
6 years, 8 months
Greek translation of el.po and a couple of questions
by Ioannis A. Venetis
Dear all,
I have completed (with the help of Manolis Tzagkarakis) the translation
for the greek language (update since 2009!!!): el.po file to gretl.mo
file.
Question what do I do with the el.po file (and/or generated gretl.mo
file)?
Just to inform everybody, an ambitious plan has just started to
translate also the command and function references!!!
While translating the el.po the following two questions/issues came up:
==============================
(A)
from console or script
the SET SEED #### command returns a comment/message in english
Pseudo-random number generator seeded with #####
although the message that appears in el.po has been translated. (same
happens with Italian or Portoguese that i have tried as a cross-check)
I changed the message in el.po from
Pseudo-random number generator seeded with %d
to
Pseudo-random number generator seeded with %u
(as it appears in libset.c) and now the translated Greek message
appears.
==============================
==============================
(B)
from console or script
SET SEED command returns
seed: unsigned int, currently 1502175471 (automatic)
which is not included in the po file for translation (ok not a big
deal!)
==============================
Thanks,
Yiannis
6 years, 8 months
