gmane.comp.statistics.gretl.user

Re: Simultaneous Equation model

Sven Schreiber — 2013-05-21T16:42:41

Am 21.05.2013 11:30, schrieb Sudipta Mahapatra:

No, IIRC in a system context nonlinear restrictions are not available in 
gretl.

sorry,
sven

Simultaneous Equation model

Sudipta Mahapatra — 2013-05-21T09:30:15

Hi,

I am new to Gretl and want to fit a set of simultaneous equations to a data
set with parameter constraints of the type: a1/b1=a3/b3. Is there a way to
do this in Gretl?

Thank you and best regards.

Sudipta Mahapatra

Re: DPANEL and Sargan/Hansen test

Pindar — 2013-05-20T10:41:57

Am 20.05.2013 11:56, schrieb Sven Schreiber:

Thanks Sven for pointing me to the 'always': The coefficients for the 
const and the time dummies differ!

Trying to change the setting for the time dummies leads to 'completely 
different' coefficients while
it does not alter the Sargan test statistic. I obviously failed in 
replicating the time dummy instruments:

<hansl>
open abdata.gdt
genr time
genr timedum
list TD_roodman = dt_2 dt_3 dt_4 dt_5 dt_6 dt_7 dt_8

dpanel 1; n const w w(-1) k k(-1) TD_roodman ; \
   GMM(n,2,8) GMM(w,2,8) GMM(k,2,8) \
   GMMlevel(w,1,1) GMMlevel(k,1,1) TD_roodman --sys

# This estimation gives a 'Sargan test'
#  Sargan over-identification test: Chi-square(100) = 154.367 [0.0004]
<hansl>

Best
Leon

Here the Roodman stata output of coefficients:

Re: DPANEL and Sargan/Hansen test

Sven Schreiber — 2013-05-20T09:56:15

Am 20.05.2013 11:52, schrieb Pindar:


Yes, a useful question I think. But are the coeff estimates always the
same, are you absolutely sure you are comparing identical
specificiations? In panel settings and GMM settings there can be subtle
differences.

cheers,
sven

Re: DPANEL and Sargan/Hansen test

Pindar — 2013-05-20T09:52:33

Am 19.05.2013 23:29, schrieb Rodrigo Alfaro Arancibia:
Hola Rodrigo,

The p-value for Hansen test is reported as " 0.218".
But with the output in the paper and gretl there are 3 different test 
statistics for chi2(100):

Sargan_xtabond2:     186.90
Sargan_gretl:             154.81
Hansen_xtabond2:   110.70

I would like to be sure how to interpret differences in the diagnostic 
checks between gretl and stata.

Leon

Re: DPANEL and Sargan/Hansen test

Rodrigo Alfaro Arancibia — 2013-05-19T21:29:49

León,

For me, such as big valué in the chi2 reject the null, whatever which one.
You should write Roodman in order to assess p-valúes.

R

El domingo, 19 de mayo de 2013, Pindar escribió:

DPANEL and Sargan/Hansen test

Pindar — 2013-05-19T16:19:04

Hi,

I'm still trying to get a feeling for the dpanel gmm estimators.
When estimating this xtabond2 statement from Roodman (2006/2008) for 
abdata.gdt
"xtabond2 n L.n L(0/1).(w k) yr*, gmmstyle(L.(n w k)) ivstyle(yr*, 
equation(level)) robust small"
     by
<hansl>
open abdata.gdt
dpanel 1; n const w w(-1) k k(-1) ; \
   GMM(n,2,8) GMM(w,2,8) GMM(k,2,8) \
   GMMlevel(w,1,1) GMMlevel(k,1,1) --time --sys
<hansl>

I came across a question concerning the Sargan/Hansen test of overid. 
restrictions.
In the gretl-guide on p.152 it is stated that "Specifically, xtabond2 
computes both a "Sargan
test" and a "Hansen test" for overidentification, but what it calls the 
Hansen test is what DPD and
gretl call the Sargan test."

The Hansen test in this example does not reject the validity of the 
instruments while the Sargan does.

"Sargan test of overid. restrictions: chi2(100) = 186.90 Prob > chi2 = 0.000
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(100) = 110.70 Prob >

Re: Gnuplot save eps

Allin Cottrell — 2013-05-19T15:04:30


I think you may need to use the no-spaces version of the font 
name, as found in its afm file. "Times-Roman-Bold,13" works 
here.

Allin Cottrell

Re: Gnuplot save eps

Gabriela Nodari — 2013-05-19T13:33:07

Done!

If it could be useful for someone else,  to use the ampersand (&) symbol in
the postscript terminal you need to escape it twice: \\&
On 19/05/2013 2:09 PM, "Gabriela Nodari" <gabriela.nodari-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org> wrote:

Re: Gnuplot save eps

Gabriela Nodari — 2013-05-19T12:09:44

Dear Allin,

I can't find a way to display "&" in the graph. When I open a saved graph,
gretl shows it, but when I save it in eps format, it does not take that
symbol...

Suggestions? Thank you very much!

Gabriela


2013/5/19 Gabriela Nodari <gabriela.nodari-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>

Re: Gnuplot save eps

Gabriela Nodari — 2013-05-19T11:10:30

Thank you very much! Now it works! I will check the help to change font
because it is not taking times bold with a large size.
Thanks again
On 19/05/2013 10:20 AM, "Allin Cottrell" <cottrell-SL7aeOxzxAQ< at >public.gmane.org> wrote:

Re: Question

Allin Cottrell — 2013-05-19T11:04:37


You cannot _define_ a function inside a loop. But you can 
call a function from a loop just fine, which I presume is what 
you really want to do.

<hansl>
function scalar trivial (scalar x)
   return 2*x
end function

# main script
loop i=1..10
   scalar k = trivial(i)
endloop
</hansl>

Allin Cottrell

Question

Antonio Elias — 2013-05-19T10:22:46

Hello all,

Sorry if my question is stupid. I am an undergraduated student and I
performaced a function with my own estimator for my final project. Now I
would like to create a monte carlo silumation for testing the behaviour of
my parameters but I read in the manual that a loop does not allow a
function inside. Would be possible do it?

Thank you very much,
Aefz

Re: IRF plot error

Allin Cottrell — 2013-05-19T09:58:44


This is likely to do with the fact that the "creator" 
information was not being recorded when a bundle was saved as 
part of a session file. That's now fixed in CVS and snapshots.

I've verified that if I create a bundle using the "IS-LM" 
example SVAR script and save the gretl session, then on 
reopening the session file the bundle has a Graph menu and IRF 
plots can be drawn.

Allin Cottrell

Re: Gnuplot save eps

Allin Cottrell — 2013-05-19T08:18:12


You're asking here for a plot that's 640 x 480 inches (the 
inch being the default unit when specifying the size of an EPS 
plot). Make that something sane and you should be OK.

More generally, in gnuplot, try "help postscript".

Allin Cottrell

IRF plot error

Gabriela Nodari — 2013-05-19T02:14:13

Dear gretl users,

I have a problem with plotting IRFs of a SVAR model saved on the icon menu.
Specifically, once I close a session (I mean closing gretl), after having
estimated and saved a SVAR model via script, I am not able to plot its IRFs
when I re-open the session. Even if the "bundle" is saved on the icon menu,
I use a script to plot the IRFs (just using IRFplot(&Mod,1,1) and I get the
following message:

gretl version 1.9.12cvs
Current session: 2013-05-19 04:06
? IRFplot(&SVARregs,2,1)
"snames": no such item
*** error in function IRFplot

Error executing script: halting

Any suggestion?

Thanks in advance.
Gabriela

Re: Gnuplot save eps

Gabriela Nodari — 2013-05-19T01:59:20

Thank you for your answer.
I got the output saved in eps format. However, it consists only of a part
of the whole graph. I don't know what is going on. Could you please take a
look to my code? The data reports IRFs results.

Please notice that I have reduce the time horizon for ease of exposition,
i.e. you will find "xrange" until 60, whereas I have reported here only 4
point estimates per IRF.

set term postscript eps "Times New Roman Bold,13" size 640,480
set encoding utf8
# user-defined plot
set style line 1 lc rgb "#0000ff"
set style line 2 lc rgb "#ff0021"
set style line 3 lc rgb "#070d07"
set style line 4 lc rgb "#e10c2c"
set style line 5 lc rgb "#00cc00"
set style line 6 lc rgb "#0000ff"
set style line 8 lc rgb "#00fff2"
set style increment user
set yrange [-11.0976:2.30634]
set xrange [0:60]
set xlabel 'months'
#set ylabel 'Industrial Production'
set xzeroaxis
set key right bottom
set title ""
set style fill solid 0.175
set output '/Users/Gabriela/Desktop/IProb.eps'
plot \
'-' using 1:2:3 notitle w f

Re: Gnuplot save eps

Riccardo (Jack) Lucchetti — 2013-05-17T19:35:19


See the help for "set terminal" and "set output". For example:

<gnuplot>
set terminal eps
set output 'Moo.eps'
plot cos(x)
</gnuplot>


-------------------------------------------------------
   Riccardo (Jack) Lucchetti
   Dipartimento di Scienze Economiche e Sociali (DiSES)

   Università Politecnica delle Marche
   (formerly known as Università di Ancona)

   r.lucchetti-kONsfx5siupeoWH0uzbU5w< at >public.gmane.org
   http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------

Re: DPANEL time dummies and the dpdstyle

Pindar — 2013-05-17T16:45:14

17.05.2013 12:38, Pindar:

Actually I wonder why this happens, and whether this is a signal for 
misspecification.


I found in Roodman (2006/2008) p. 42 the necessary "gynmastics" for 
'Difference GMM' in stat code:

This example exactly imitates the regression for column (a1), Table 4 in 
Arellano and Bond (1991):

forvalues y = 1979/1984 { /* Make variables whose differences are time 
dummies */
gen yr`y?c = year >= `y?
}
gen cons = year
xtabond2 n L(0/1).(L.n w) L(0/2).(k ys) yr198?c cons, gmm(L.n) 
iv(L(0/1).w L(0/2).(k ys)
 > yr198?c cons) noleveleq noconstant small robust

What is he doing here? This is what the --dpd flag does, right?
I detect the the loop, but don't know how to adapt the code in HANSL.
(I have no experience with STATA code)

Gnuplot save eps

Gabriela Nodari — 2013-05-17T16:39:26

Dear gretl users,

I have modified the code of a graph generated by gretl. I am no more able
to save this graph in eps format using menu options. Does someone know what
I should write in the code to save this graph?

Thanks in advance!
Gabriela

Re: restrict --silent option

Allin Cottrell — 2013-05-17T14:47:58


Up till now, the --full flag was being taken as disabling 
--silent, but now in CVS --silent is respected regardless.

Allin Cottrell