2:37 a.m.
In an earlier post (with a different subject line) I mentioned that
the SVAR graphs could be more informative by adding "IRF by shock" and
IRF by var". That same principle might apply to adding the model
option to the graph title, e.g., "Plain model" or "KPSW Model."
And at a higher level of menu items: if I understand the SVAR
documentation correctly (and the possibility that I do not is quite
high), the KPSW Model of SVAR is actually a SVEC? If so, then to a
naive user at least (e.g., me) moving that option off of the SVAR and
into its own sub-menu-item of Time Series would make sense.. call it
SVEC (KPSW) perhaps.. Thanks TMN
9:19 p.m.
On Fri, 2 May 2014, Tim Nall wrote:
In an earlier post (with a different subject line) I mentioned that
the SVAR graphs could be more informative by adding "IRF by shock" and
IRF by var". That same principle might apply to adding the model
option to the graph title, e.g., "Plain model" or "KPSW Model."
Yes, that could be done. I'll give it a good look next week. However, I'm
a bit reluctant to give "official" status to the terms "plain" and
"KPSW",
as they're non-standard in the literature, and I just used them in the
context of my SVAR package as working expressions. And bear in mind that
graphs are editable, so if you like those phrases (which I don't very
much), you're welcome to add them in your own work.
Moreover, there's a more general thing I'd like to point out. The SVAR
package is meant to be used primarily via scripts. The GUI interface must
be considered a convenient shortcut for some of its basic features, but
the specification, estimation and analysis of a SVAR model are too complex
to be handled through a GUI to a satisfactory degree.
I don't plan to spend too much time working on the GUI interface to SVAR:
I'm simply not interested in that. I'd much prefer devote my time and
energy in debugging and extending its number-crunching part; for example,
I'd like to incorporate the algorithm for assessing identification,
sign-identified models à la Uhlig, heteroskedasticity-identified models à
la Rigobon with Bacchiocchi's generalisation... there's so much more that
could be done.
If you fell like spending time perfecting the GUI to it. you're very
welcome, and I'll be happy to offer all the help I can.
And at a higher level of menu items: if I understand the SVAR
documentation correctly (and the possibility that I do not is quite
high), the KPSW Model of SVAR is actually a SVEC? If so, then to a
naive user at least (e.g., me) moving that option off of the SVAR and
into its own sub-menu-item of Time Series would make sense.. call it
SVEC (KPSW) perhaps.. Thanks TMN
Yes, up to a point. A VECM is, from a statistical viewpoint, a particolar
subspecies of a VAR in which we assume that we know how many unit roots
are in the system so that a very clever system of constraints can be put
on the VAR representation. This has no implication on the "structural"
part _per se_. You can use a triangular (or Cholesky, if you prefer, or
recursive, if you prefer) identification scheme on a VECM and it will
become a SVECM. What *I* call a KPSW model is a model in which you taker
the extra step of using your assumed knowledge of the cointegration space
to classify structural shocks into permanent and transitory. So a KPSW
model is a SVECM, but the converse doesn't necessarily hold.
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------
10:04 p.m.
Dear Riccardo,
I'm am going through SVAR.pdf right now I don't know why, but my version (cvs
compiled 2 days ago) doesn't eat horisont I put for control Mod["horizon"] =
40 x=Mod["horizon"] x output is ok: ? x 40 but IRFplot(&Mod, 1, 1)
gives default 20 lags
--- Оригінальне повідомлення ---
Від кого: "Riccardo (Jack) Lucchetti" < r.lucchetti(a)univpm.it >
Дата: 3 травня 2014, 00:25:15
On Fri, 2 May 2014, Tim Nall wrote:
In an earlier post (with a different subject line) I mentioned that
the SVAR graphs could be more informative by adding "IRF by shock" and
IRF by var". That same principle might apply to adding the model
option to the graph title, e.g., "Plain model" or "KPSW Model."
Yes, that could be done. I'll give it a good look next week. However, I'm
a bit reluctant to give "official" status to the terms "plain" and
"KPSW",
as they're non-standard in the literature, and I just used them in the
context of my SVAR package as working expressions. And bear in mind that
graphs are editable, so if you like those phrases (which I don't very
much), you're welcome to add them in your own work.
Moreover, there's a more general thing I'd like to point out. The SVAR
package is meant to be used primarily via scripts. The GUI interface must
be considered a convenient shortcut for some of its basic features, but
the specification, estimation and analysis of a SVAR model are too complex
to be handled through a GUI to a satisfactory degree.
I don't plan to spend too much time working on the GUI interface to SVAR:
I'm simply not interested in that. I'd much prefer devote my time and
energy in debugging and extending its number-crunching part; for example,
I'd like to incorporate the algorithm for assessing identification,
sign-identified models à la Uhlig, heteroskedasticity-identified models à
la Rigobon with Bacchiocchi's generalisation... there's so much more that
could be done.
If you fell like spending time perfecting the GUI to it. you're very
welcome, and I'll be happy to offer all the help I can.
And at a higher level of menu items: if I understand the SVAR
documentation correctly (and the possibility that I do not is quite
high), the KPSW Model of SVAR is actually a SVEC? If so, then to a
naive user at least (e.g., me) moving that option off of the SVAR and
into its own sub-menu-item of Time Series would make sense.. call it
SVEC (KPSW) perhaps.. Thanks TMN
Yes, up to a point. A VECM is, from a statistical viewpoint, a particolar
subspecies of a VAR in which we assume that we know how many unit roots
are in the system so that a very clever system of constraints can be put
on the VAR representation. This has no implication on the "structural"
part _per se_. You can use a triangular (or Cholesky, if you prefer, or
recursive, if you prefer) identification scheme on a VECM and it will
become a SVECM. What *I* call a KPSW model is a model in which you taker
the extra step of using your assumed knowledge of the cointegration space
to classify structural shocks into permanent and transitory. So a KPSW
model is a SVECM, but the converse doesn't necessarily hold.
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------
_______________________________________________
Gretl-users mailing list
Gretl-users(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users
3:44 p.m.
On Sat, 3 May 2014, oleg_komashko(a)ukr.net wrote:
Dear Riccardo,
I'm am going through SVAR.pdf right now I don't know why, but
my version
(cvs compiled 2 days ago) doesn't eat horisont I put for control
Mod["horizon"] = 40 x=Mod["horizon"] x output is ok: ? x 40 but
IRFplot(&Mod, 1, 1)
gives default 20 lags
Are you sure you're setting the "horizon" scalar _before_ calling
SVAR_estimate?
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------
5:16 p.m.
Thank you, I will try now. I've only copy-pasted from SVAR.pdf Yes! Graphics has
become much nicer.
But I'd never guessed the right order of commands if without your letter. sprintf
still doesn' work in Linux. I'v got the chart by IRFplot(&Mod, i, j)
P.S. I have seen your 2006 article long before I started using gretl when I prepared a
textbook chapter. So I well understand your level and guess you are interested in really
cool things to implement in gretl. I feel slightly guilty to torture you with tiny
bugs. Two things to cheer up. I am planning to enlist to gretl about 200 new students
every year. Imagine potential for gretl users. Once, using google translator I looked for
some Chinese gretl and econometric references for my students from China. I found a
Chinese language site with a counter of gretl users: 8000 Taiwan 2000 Mainland. What will
be, when the rules against piracy in the mainland China will become civilized. A don't
know, may be it's blocked in the civilized world, but the web is swollen with
perfectly cracked e-viewses and statas. When I changed to Linux, R and gretl, everybody
looked at me as at a freak.
--- Оригінальне повідомлення ---
Від кого: "Riccardo (Jack) Lucchetti" < r.lucchetti(a)univpm.it >
Дата: 4 травня 2014, 18:44:08
On Sat, 3 May 2014, oleg_komashko(a)ukr.net wrote:
Dear Riccardo,
I'm am going through SVAR.pdf right now I don't know why, but
my version
(cvs compiled 2 days ago) doesn't eat horisont I put for control
Mod["horizon"] = 40 x=Mod["horizon"] x output is ok: ? x 40 but
IRFplot(&Mod, 1, 1)
gives default 20 lags
Are you sure you're setting the "horizon" scalar _before_ calling
SVAR_estimate?
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------
5:31 p.m.
On Sun, 4 May 2014, oleg_komashko(a)ukr.net wrote:
Thank you, I will try now. I've only copy-pasted from SVAR.pdf
Yes! Graphics has become much nicer.
But I'd never guessed the right order of commands if without your letter.
Well, it _is_ mentioned in the SVAR documentation...
sprintf still doesn' work in Linux. I'v got the chart
by IRFplot(&Mod, i, j)
???
P.S. I have seen your 2006 article long before I started using gretl
when I prepared a textbook chapter. So I well understand your level and
guess you are interested in really cool things to implement in gretl. I
feel slightly guilty to torture you with tiny bugs.
Thank you. If you're interested in SVARs, you can contribute too. As
you're probably aware of by now, gretl's SVAR module is written entirely
in hansl (source at
http://gretl.cvs.sourceforge.net/viewvc/gretl/gretl-addons/SVAR/). If you
feel like improving on it (and believe me, there are lots of areas that
could benefit from some extra work), you're welcome to it! :)
Two things to cheer up. I am planning to enlist to gretl about 200
new
students every year. Imagine potential for gretl users. Once, using
google translator I looked for some Chinese gretl and econometric
references for my students from China. I found a Chinese language site
with a counter of gretl users: 8000 Taiwan 2000 Mainland. What will be,
when the rules against piracy in the mainland China will become
civilized. A don't know, may be it's blocked in the civilized world, but
the web is swollen with perfectly cracked e-viewses and statas. When I
changed to Linux, R and gretl, everybody looked at me as at a freak.
It's understandable that many people start using free software because you
don't have to pay for it. Our goal, however, is to provide a better tool
than commercial software, not just with a better quality/price ratio. In
some areas, I'd venture to say that our implementation of certain
statistical techniques is on par, if not better, than any other package
(commercial or not); in some other area, R is king. Of course, in some
other areas we're behind, no question about it. For example, I still think
that the greatest strength of Stata is the quality of their manuals. I
wish we had manuals comparable to theirs. Oh, well.
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------
6:27 p.m.
Thank you for your suggestion. I will try. In research I am more macroeconomic-oriented,
so it's the bull's eye .
"Historically" I accustomed to f[___]:=Module not to --xxx-yyy but it is quick
to overcome, I hope
I have PhD in probability. I'm proud, http://en.wikipedia.org/wiki/Anatoliy_Skorokhod
was one of my referees. That was established school. Even my graduate diploma was
translated in USA. But for a Ukrainian economist, it is very difficult to get out from
obscurity.
--- Оригінальне повідомлення ---
Від кого: "Riccardo (Jack) Lucchetti" < r.lucchetti(a)univpm.it >
Дата: 4 травня 2014, 20:31:35
On Sun, 4 May 2014, oleg_komashko(a)ukr.net wrote:
Thank you, I will try now. I've only copy-pasted from SVAR.pdf
Yes! Graphics has become much nicer.
But I'd never guessed the right order of commands if without your letter.
Well, it _is_ mentioned in the SVAR documentation...
sprintf still doesn' work in Linux. I'v got the chart
by IRFplot(&Mod, i, j)
???
P.S. I have seen your 2006 article long before I started using gretl
when I prepared a textbook chapter. So I well understand your level and
guess you are interested in really cool things to implement in gretl. I
feel slightly guilty to torture you with tiny bugs.
Thank you. If you're interested in SVARs, you can contribute too. As
you're probably aware of by now, gretl's SVAR module is written entirely
in hansl (source at
http://gretl.cvs.sourceforge.net/viewvc/gretl/gretl-addons/SVAR/ ). If you
feel like improving on it (and believe me, there are lots of areas that
could benefit from some extra work), you're welcome to it! :)
Two things to cheer up. I am planning to enlist to gretl about 200
new
students every year. Imagine potential for gretl users. Once, using
google translator I looked for some Chinese gretl and econometric
references for my students from China. I found a Chinese language site
with a counter of gretl users: 8000 Taiwan 2000 Mainland. What will be,
when the rules against piracy in the mainland China will become
civilized. A don't know, may be it's blocked in the civilized world, but
the web is swollen with perfectly cracked e-viewses and statas. When I
changed to Linux, R and gretl, everybody looked at me as at a freak.
It's understandable that many people start using free software because you
don't have to pay for it. Our goal, however, is to provide a better tool
than commercial software, not just with a better quality/price ratio. In
some areas, I'd venture to say that our implementation of certain
statistical techniques is on par, if not better, than any other package
(commercial or not); in some other area, R is king. Of course, in some
other areas we're behind, no question about it. For example, I still think
that the greatest strength of Stata is the quality of their manuals. I
wish we had manuals comparable to theirs. Oh, well.
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------
6:05 p.m.
New subject: on contributing SVAR/SVEC
Riccardo, On trying to reproduce output from awm I have learned how to estimate SVEC in R
and jMulTi. If I understand correctly I can start by playing with
http://gretl.cvs.sourceforge.net/viewvc/gretl/gretl-addons/SVAR/SVAR.inp?... ,
Revision 1.15.
Am I right that function matrix SVECM(list X, list W, scalar varorder, scalar h, matrix
Rd, matrix beta) is planned for KRSW but you haven't enough time to polish it? Can you
suggest something not very difficult for a start. Such things obviously can spare your
time for something more advanced.
P.S. Have read about such approach. Run computeable DGE. Generate irfs. Estimate
corresponding VAR. Introduce a metric for irfs. Choose ifrs for estamated VAR with the
closest distance to DGE's irf's. Brrrr! Such irfs will never be realistic and will
not be general in Friedman's sense.
--- Оригінальне повідомлення ---
Від кого: "Riccardo (Jack) Lucchetti" < r.lucchetti(a)univpm.it >
Дата: 4 травня 2014, 20:31:35
On Sun, 4 May 2014, oleg_komashko(a)ukr.net wrote:
Thank you, I will try now. I've only copy-pasted from SVAR.pdf
Yes! Graphics has become much nicer.
But I'd never guessed the right order of commands if without your letter.
Well, it _is_ mentioned in the SVAR documentation...
sprintf still doesn' work in Linux. I'v got the chart
by IRFplot(&Mod, i, j)
???
P.S. I have seen your 2006 article long before I started using gretl
when I prepared a textbook chapter. So I well understand your level and
guess you are interested in really cool things to implement in gretl. I
feel slightly guilty to torture you with tiny bugs.
Thank you. If you're interested in SVARs, you can contribute too. As
you're probably aware of by now, gretl's SVAR module is written entirely
in hansl (source at
http://gretl.cvs.sourceforge.net/viewvc/gretl/gretl-addons/SVAR/ ). If you
feel like improving on it (and believe me, there are lots of areas that
could benefit from some extra work), you're welcome to it! :)
Two things to cheer up. I am planning to enlist to gretl about 200
new
students every year. Imagine potential for gretl users. Once, using
google translator I looked for some Chinese gretl and econometric
references for my students from China. I found a Chinese language site
with a counter of gretl users: 8000 Taiwan 2000 Mainland. What will be,
when the rules against piracy in the mainland China will become
civilized. A don't know, may be it's blocked in the civilized world, but
the web is swollen with perfectly cracked e-viewses and statas. When I
changed to Linux, R and gretl, everybody looked at me as at a freak.
It's understandable that many people start using free software because you
don't have to pay for it. Our goal, however, is to provide a better tool
than commercial software, not just with a better quality/price ratio. In
some areas, I'd venture to say that our implementation of certain
statistical techniques is on par, if not better, than any other package
(commercial or not); in some other area, R is king. Of course, in some
other areas we're behind, no question about it. For example, I still think
that the greatest strength of Stata is the quality of their manuals. I
wish we had manuals comparable to theirs. Oh, well.
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------
3928
days inactive
3933
days old
7 comments
3 participants
participants (3)
-
oleg_komashko＠ukr.net -
Riccardo (Jack) Lucchetti -
Tim Nall