VECM, Impulse response: gretl crashes
by Artur T.
Hello,
I just estimated a large VECM model. Gretl crashes when I run impulse
response functions with bootstrapped confidence intervals. Computing
impulse response functions without these works fine. I attached the
gretl session file where you can find model number 1 saved.
Hopefully, the issue can be solved.
Thanks your effort.
Best,
Artur
14 years, 10 months
Re: [Gretl-users] GRETL: Return Computation
by Kehl D ániel
Dear Simari,
try the following short script:
<script>
list stocks = dataset
loop foreach i stocks
genr r_$i = ($i-$i(-1))/$i(-1)*100
endloop
<\script>
HTH:
Daniel
-----Original Message-----
From: Simari <gnsimari(a)gmail.com>
To: Gretl list <gretl-users(a)lists.wfu.edu>
Date: Sat, 20 Mar 2010 11:34:01 +0100
Subject: [Gretl-users] GRETL: Return Computation
Dear,
anyone knows how to compute the return of all stocks (500) in only one
command?
I have to insert this formula
"(Microsft-Microsoft(-1))/Micorsoft(-1)*100
for 500 different stocks and I would like to know if there's a way to do
that giving only one command.
Thanks,
G. Simari
On Fri, Mar 19, 2010 at 5:54 PM, Allin Cottrell <cottrell(a)wfu.edu> wrote:
>
> On Fri, 19 Mar 2010, Talha Yalta wrote:
>
> > After creating a scatter-plot, I go into plot controls, select add
> > line and enter "2*x". Gretl crashes as soon as I select OK or Close.
>
> Thanks, that's now fixed in CVS and snapshots.
>
> > Also, I noticed during translation the entry regarding confidence
> > bands. How do I reach this function? Is this available for a regular
> > OLS regression?
>
> I'm not quite sure what you're asking. You can get a confidence
> band for forecasts from OLS models (as you probably know). What's
> new is a choice of graphical representation of the confidence
> interval for bootstrapped VAR impulse responses.
>
> Allin Cottrell
>
>
> _______________________________________________
> Gretl-users mailing list
> Gretl-users(a)lists.wfu.edu
> http://lists.wfu.edu/mailman/listinfo/gretl-users
>
14 years, 10 months
plot gui error
by Talha Yalta
Hi:
After creating a scatter-plot, I go into plot controls, select add
line and enter "2*x". Gretl crashes as soon as I select OK or Close.
Here is the console output:
(gretl_x11:11396): Gtk-CRITICAL **: gtk_spin_button_get_value_as_int:
assertion `GTK_IS_SPIN_BUTTON (spin_button)' failed
get_png_bounds_info(): failed
*** glibc detected *** /opt/gretl/bin/gretl_x11: double free or
corruption (fasttop): 0x096d3538 ***
Also, I noticed during translation the entry regarding confidence
bands. How do I reach this function? Is this available for a regular
OLS regression?
Cheers
Talha
--
“Remember not only to say the right thing in the right place, but far
more difficult still, to leave unsaid the wrong thing at the tempting
moment.” - Benjamin Franklin (1706-1790)
--
14 years, 10 months
Gretl and MacOs 10.4.11
by Simari
Dear,
I use MacOs 10.4.11 (a Mac Book)
I would like to install Gretl, becausa I started using Econometrics.
I tried to download the software but it didn't work.
Can you help me?
Regards,
G. Simari
14 years, 10 months
Gretl Guide: Error message
by Henrique Andrade
Dear Gretl Community,
I'm trying to open Gretl User's Guide (or Command reference), Help -> User's
guide, and I'm getting this error message:
"C:\Documents and Settings\henrique\Dados de
aplicativos\gretl\doc\gretl-guide-a4.pdf: No such file or directory"
If I select Tools -> Preferences -> General -> Manual and check English (US
letter paper) in the PDF manual preference the problem disappears. This
problem only happens in Windows XP (no problems in Vista or Mac OS/X).
Best,
Henrique C. de Andrade
Doutorando em Economia Aplicada
Universidade Federal do Rio Grande do Sul
www.ufrgs.br/ppge
14 years, 10 months
VAR Session Error
by Henrique Andrade
Dear Gretl Community,
I saved a session with a VAR (with only one lag) and when I try to reopen
that session the following error message appears: "1 session object(s) could
not be rebuilt". If I include different lag durations only the 1-lagged VAR
can't be rebuild. This problems only happens when I'm using Gretl for
Windows (XP or Vista). In Gretl for Mac I don't have problems.
Best,
Henrique C. de Andrade
Doutorando em Economia Aplicada
Universidade Federal do Rio Grande do Sul
www.ufrgs.br/ppge
14 years, 10 months
Ralph M Rodriguez/PO/KAIPERM is out of the office.
by Ralph.M.Rodriguez＠kp.org
I will be out of the office starting 03/17/2010 and will not return until
03/22/2010.
Hi All, I will be out of the offcie from March 17 returning to the office
on Monday March 22, 2010. Please, if you have an immediate question
regarding Cost Model or Construction Economics, please contact Brad A Njus,
510 625 4595, with your questions.
Ralph
14 years, 10 months
hp confidence interval building
by Patricio Cuarón
Hello. I'm trying to implement the Gallego-Johnson method for building
confidence intervals for the HP filter. The methodology is explained in this
diagram:
I wrote the following code:
smpl full
genr gdp_d11 = pbip_d11
genr cycle = hpfilt(gdp_d11)
genr trend = gdp_d11-cycle
matrix A = {}
matrix A = A ~ {cycle}
scalar n = 1000
loop for counter=1..n --quiet
genr auxvar1=trend+resample(cycle,22)
genr auxvar2=auxvar1-hpfilt(auxvar1)
genr auxvar3 = hpfilt(auxvar1)
matrix A = A ~ {auxvar3}
endloop
>
> mubgap=quantile(transp(A),0.95)
mlbgap=quantile(transp(A),0.05)
> series ubgap=mubgap
series lbgap=mlbgap
> genr ubgap=ubgap+cycle
genr lbgap=cycle-lbgap
But the outputs are wrong. I don't get what I'm doing wrong. I'd be grateful
for some help.
Thanks!
PC
14 years, 10 months
VAR: Impulse-response
by Henrique Andrade
Dear Gretl Community,
I have a question about how Gretl calculates the VAR
impulse-responses. Although
Gretl says that an one-standard error shock is used in the calculations I
can't confirm this, once another value for the standard error is given in
the estimation results. To illustrate this I will give an example:
<script>
open australia.gdt
"Modelo 1" <- var 1 le lpus lpau --impulse-responses
</script>
The standard-error of the "Equation 1" (le) is 0.048110, but the
impulse-response uses 0.046827 as the standard-error. I did the same
estimation using EViews and, according to that software, 0.046827 is the
standard-error with degrees of freedom correction and 0.046827 is the
standard-error without degrees of freedom correction.
Best,
Henrique C. de Andrade
Doutorando em Economia Aplicada
Universidade Federal do Rio Grande do Sul
www.ufrgs.br/ppge
14 years, 10 months
list question
by Kehl D ániel
Dear List-members,
I have a really simple question.
I want to compress my small code below:
list ylist = 10 13 16 19
loop foreach i ylist
series ln2_$i = ln(ylist.$i*GERD95)
endloop
list ylist = 11 14 17 20
loop foreach i ylist
series ln2_$i = ln(ylist.$i*GERD97)
endloop
list ylist = 12 15 18 21
loop foreach i ylist
series ln2_$i = ln(ylist.$i*GERD99)
endloop
How is it possible to make mylist dynamic in the sense i can add +1 to the ID's in it.
By the way, with GERDXX the case is pretty the same.
Thanks:
Daniel
14 years, 10 months
