Sven Schreiber | 23 Jul 14:17 2014
Picon
Picon

Variable generation full vs. restricted sample

Hi,

obviously it's absolutely essential to be able to create variables only 
for the currently active subsample. But I'm wondering, is there another 
(easier) way to generate variables also for the full workfile sample 
range without temporarily removing and then later re-applying the sample 
restrictions?

Perhaps something like "series mynew = log(income) --full" if you 
understand what I mean by that.

thanks for suggestions,
sven
Logan Kelly | 19 Jul 21:28 2014

Question about SVAR

Hello,

 

I am estimating a “plain” model with the SVAR package  because the native irf() function returns :

 

Matrix is not positive definite

 

(Note: the reason for this error is discussed in another post)

 

Thus, I am trying the SVAR package, but the “actual” irf estimate is always above the bootstraped confidence bands. I am using the bias corrected bootstrapping method (Killian 1989).

 

I am thinking this must be a problem with my data, not SVAR or gretl? Does that sound right?

 

SVAR: 0.997

gretl:  1.9.90

os: win 7 64 bit

 

Thanks,

 

Logan

<div>
<div class="WordSection1">
<p class="MsoNormal">Hello,<p></p></p>
<p class="MsoNormal"><p>&nbsp;</p></p>
<p class="MsoNormal">I am estimating a &ldquo;plain&rdquo; model with the SVAR package&nbsp; because the native irf() function returns :<p></p></p>
<p class="MsoNormal"><p>&nbsp;</p></p>
<p class="MsoNormal">Matrix is not positive definite<p></p></p>
<p class="MsoNormal"><p>&nbsp;</p></p>
<p class="MsoNormal">(Note: the reason for this error is discussed in another post)<p></p></p>
<p class="MsoNormal"><p>&nbsp;</p></p>
<p class="MsoNormal">Thus, I am trying the SVAR package, but the &ldquo;actual&rdquo; irf estimate is always above the bootstraped confidence bands. I am using the bias corrected bootstrapping method (Killian 1989).<p></p></p>
<p class="MsoNormal"><p>&nbsp;</p></p>
<p class="MsoNormal">I am thinking this must be a problem with my data, not SVAR or gretl? Does that sound right?<p></p></p>
<p class="MsoNormal"><p>&nbsp;</p></p>
<p class="MsoNormal">SVAR: 0.997<p></p></p>
<p class="MsoNormal">gretl: &nbsp;1.9.90<p></p></p>
<p class="MsoNormal">os: win 7 64 bit<p></p></p>
<p class="MsoNormal"><p>&nbsp;</p></p>
<p class="MsoNormal">Thanks,<p></p></p>
<p class="MsoNormal"><p>&nbsp;</p></p>
<p class="MsoNormal">Logan<p></p></p>
</div>
</div>
Sven Schreiber | 18 Jul 18:13 2014
Picon
Picon

accessors for dpanel

Hi,

is there a way to store some of the diagnostic output from the 'dpanel'
command? For example, the '$sargan' accessor doesn't work with 'dpanel'.
Other useful retrievals could be the autocorrelation test results.

And with respect to the system variant of dpanel: Perhaps it would be
useful to automatically test the additional moment conditions, by
comparing the differences-only and the system specification?

thanks,
sven

JOSE FRANCISCO PERLES RIBES | 10 Jul 18:48 2014
Picon

Gretl versus Eviews when testing after ARIMA or AR models

Dear list:

First of all, sorry if the question is basic.

I'm exploring the issue of modelling non-linear time series. I have read in several articles that a correct strategy is often to start by fitting a linear model (for example a simple Autoregressive model) and if it is not satisfactory, then try to fit a TAR, SETAR or Markov Switching Model,.... I think this is a classical approach.

In the papers that I have read, in order to detect deviation from linearity most authors apply several test over the residuals of the simple first estimated AR model (for example RESET test, BDS test, Mc. Leod test are common in this context). Then after reject the Null in these test, they proceed with the non-linear model.

Playing with some series in Gretl, I have seen that after estimating an AR(1) model or ARIMA model with the options built in Gretl-GUI, the window "test" post-estimation option only allows to test for Normality or ARCH, but other options as RESET or Non-linearity are not activated.

Only after estimating a model via OLS menu with the dependent variable and its lags -that is not exactly  the same of estimating the AR(1) model, so the constant change as usual- these post-estimation options are allowed.

I have checked that in Eviews, for example, the options RESET, etc. are (like not in Gretl) allowed after estimating both, the AR(1) model and the equation with the lagged dependent variable.

Y C AR(1)  AR residual term model
Y C Y(-1) model with the lagged endogenous variable

So, my question is: Why this difference among Gretl and Eviews, the disallowed options in Gretl are for some special considerations?

If I want to perform this kind of analysis in Gretl, with an AR model, which is the correct form to proceed?. Save the residuals of my estimated AR model and export them to R or other software to perform the BDS or RESET test as usual in literature?

Thanks in advance.

José Perles
University of Alicante
Spain
<div><div dir="ltr">
<div>
<div>
<div>Dear list:<br><br>
</div>First of all, sorry if the question is&nbsp;basic. <br><br>
</div>I'm exploring the issue of modelling non-linear time series. I have read in several articles that a correct strategy is often to start by fitting a linear model (for example a simple Autoregressive model) and if it is not satisfactory, then try to fit a TAR, SETAR or Markov Switching Model,.... I think this is a classical approach.<br><br>
</div>
<div>In the papers that I have read, in order to detect deviation from linearity most authors apply several test over the residuals of the simple first estimated AR model (for example RESET test, BDS test, Mc. Leod test are common in this context). Then after reject the Null in these test, they proceed with the non-linear model. <br><br>
</div>
<div>Playing with some series in Gretl, I have seen that after estimating an AR(1) model or ARIMA model with the options built in Gretl-GUI, the window "test" post-estimation option only allows to test for Normality or ARCH, but other options as RESET or Non-linearity are not activated. <br><br>
</div>
<div>Only after estimating a model via OLS menu with the dependent variable and its lags -that is not exactly &nbsp;the same of estimating the AR(1) model, so the constant change as usual- these post-estimation options are allowed.</div>
<div><br></div>
<div>I have checked that in Eviews, for example,&nbsp;the options RESET, etc. are (like not in Gretl) allowed after estimating both, the AR(1) model and the equation with the lagged dependent variable.<br><br>
</div>
<div>Y C AR(1)&nbsp; AR residual term model <br>
</div>
<div>Y C Y(-1) model with the lagged endogenous variable <br>
</div>
<div><br></div>
<div>So, my question is: Why this difference among Gretl and Eviews, the disallowed options in Gretl are for some special considerations?<br>
</div>
<div><br></div>
<div>If I want to perform this kind of analysis in Gretl, with an AR model, which is the correct form to proceed?. Save the residuals of my estimated AR model and export them to R or other software to perform the BDS or RESET test as usual in literature?<br>
</div>
<div><br></div>
<div>Thanks in advance.<br><br>
</div>
<div>Jos&eacute; Perles<br>
</div>
<div>University of Alicante<br>
</div>
<div>Spain</div>
</div></div>
henrique.andrade | 9 Jul 21:34 2014
Picon

Automatic "X-12-ARIMA" Specification

Dear Gretl Community,

I would like to know how to reproduce an automatic X-12-ARIMA specification inside Gretl. I'll try to explain better...

<hansl>
open fedstl.bin
data paynsa

smpl 2000:01
</hansl>

Using the GUI facilities to X-12-ARIMA
(Menu -> Variable -> X-12-ARIMA Analysis) I get the following specification:

"Final automatic model choice : (0 2 1)(0 1 1)"

And I get the following model:

<model>
Estimation converged in   10 ARMA iterations,   31 function evaluations.
 ARIMA Model:  (0 2 1)(0 1 1)
   Nonseasonal differences: 2
   Seasonal differences:    1
                                              Standard
 Parameter                    Estimate          Errors
 -----------------------------------------------------
 Nonseasonal MA                                   
   Lag  1                       0.4834         0.06713

 Seasonal MA                                      
   Lag 12                       0.7977         0.05225
</model>

So I use it with the "arima" command:

<hansl>
arima 0 2 1 ; 0 1 1 ; paynsa --nc --x-12-arima
</hansl>

This gives me the following result:

<model>
Modelo 3: ARIMA, usando as observações 2000:01-2014:04 (T = 172)
Estimado usando X-12-ARIMA (Máxima verossimilhança exata)
Variável dependente: (1-L)^2(1-Ls) paynsa

             coeficiente   erro padrão      z       p-valor
  ----------------------------------------------------------
  theta_1     -0,545388     0,0619664     -8,801   1,35e-018 ***
  Theta_1     -0,789758     0,0506616    -15,59    8,66e-055 ***
</model>

How can I exactly reproduce the model estimated by the X-12-ARIMA procedure?

Best regards,
Henrique Coêlho de Andrade

Diretoria de Estratégia e Organização

Divisão de Cenários e Estudos Macroeconômicos

Banco do Brasil

henrique.andrade-imD/zwYkNalQFI55V6+gNQ@public.gmane.org

<div> <span>Dear Gretl Community,<br><br>I would like to know how to reproduce an automatic X-12-ARIMA specification inside Gretl. I'll try to explain better...<br><br>&lt;hansl&gt;<br>open fedstl.bin<br>data paynsa<br><br>smpl 2000:01<br></span><span><span>&lt;/hansl&gt;<br><br></span>Using the GUI facilities to X-12-ARIMA </span><span><span>(Menu -&gt; Variable -&gt; X-12-ARIMA Analysis) </span>I get the following specification:<br><br>"Final automatic model choice : (0 2 1)(0 1 1)"<br><br>And I get the following model:<br><br>&lt;model&gt;<br>Estimation converged in&nbsp;&nbsp; 10 ARMA iterations,&nbsp;&nbsp; 31 function evaluations.<br>&nbsp;ARIMA Model:&nbsp; (0 2 1)(0 1 1)<br>&nbsp;&nbsp; Nonseasonal differences: 2<br>&nbsp;&nbsp; Seasonal differences:&nbsp;&nbsp;&nbsp; 1<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Standard<br>&nbsp;Parameter&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Estimate&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Errors<br>&nbsp;-----------------------------------------------------<br>&nbsp;Nonseasonal MA&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <br>&nbsp;&nbsp; Lag&nbsp; 1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.4834&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.06713<br><br>&nbsp;Seasonal MA&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <br>&nbsp;&nbsp; Lag 12&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.7977&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.05225<br></span><span><span>&lt;/model&gt;<br><br></span>So I use it with the "arima" command:<br><br>&lt;hansl&gt;<br>arima 0 2 1 ; 0 1 1 ; paynsa --nc --x-12-arima<br></span><span><span>&lt;/hansl&gt;</span><br><br>This gives me the following result:<br><br>&lt;model&gt;<br>Modelo 3: ARIMA, usando as observa&ccedil;&otilde;es 2000:01-2014:04 (T = 172)<br>Estimado usando X-12-ARIMA (M&aacute;xima verossimilhan&ccedil;a exata)<br>Vari&aacute;vel dependente: (1-L)^2(1-Ls) paynsa<br><br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; coeficiente&nbsp;&nbsp; erro padr&atilde;o&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; z&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; p-valor <br>&nbsp; ----------------------------------------------------------<br>&nbsp; theta_1&nbsp;&nbsp;&nbsp;&nbsp; -0,545388&nbsp;&nbsp;&nbsp;&nbsp; 0,0619664&nbsp;&nbsp;&nbsp;&nbsp; -8,801&nbsp;&nbsp; 1,35e-018 ***<br>&nbsp; Theta_1&nbsp;&nbsp;&nbsp;&nbsp; -0,789758&nbsp;&nbsp;&nbsp;&nbsp; 0,0506616&nbsp;&nbsp;&nbsp; -15,59&nbsp;&nbsp;&nbsp; 8,66e-055 ***<br>&lt;/model&gt;<br><br>How can I exactly reproduce the model estimated by the X-12-ARIMA procedure?<br><br>Best regards,<br><span class="Apple-style-span"><div>Henrique Co&ecirc;lho de Andrade<span class="Apple-style-span"><div><span><span></span></span></div></span><span class="Apple-style-span"><span class="Apple-style-span"><span><p>Diretoria de Estrat&eacute;gia e Organiza&ccedil;&atilde;o</p>
<p>Divis&atilde;o de Cen&aacute;rios e Estudos Macroecon&ocirc;micos</p>
<p>Banco do Brasil</p>
<p>henrique.andrade@...<br></p></span></span></span>
</div></span></span>
</div>
Artur T. | 10 Jul 12:15 2014

problems with gnuplot

Hi all,

I am currently using gretl cvs 1.9.91 on Windows 8. I set up the following example to illustrate that gnuplot does not recognize my color choice, as the two time-series still use the default color scheme butnot mine. The gnuplot version is 5.0 rc1.

<hansl>
set echo off
set messages off

open denmark.gdt --quiet
# Print TSs:
gnuplot LRM LRY --with-lines --time-series \
--output="C:\Users\artur.tarassow\Desktop\TEST.pdf" \
{ set terminal pdfcairo font 'Helvetica,15' lw 3 ; \
set style line 1 lt 1 lc rgb 'black' ; \
set style line 2 lt 2 lc rgb 'green' ; \
set style line 3 lt 2 lc rgb 'red' ; \
set style line 4 lt 2 lc rgb 'green' ; \
set style line 5 lt 2 lc rgb 'orange' ; \
set style line 6 lt 6 lc rgb 'black' ; \
set ylabel '' ; set xlabel 'Horizon' ; \
set key bottom below ; }
</hansl>

Artur
<div><div dir="ltr">
<div>
<div>
<div>
<div>Hi all,<br><br>
</div>I am currently using gretl cvs 1.9.91 on Windows 8. I set up the following example to illustrate that gnuplot does not recognize my color choice, as the two time-series still use the default color scheme butnot mine. The gnuplot version is 5.0 rc1.<br><br>&lt;hansl&gt;<br>
</div>set echo off<br>set messages off<br><br>open denmark.gdt --quiet<br># Print TSs:<br>gnuplot LRM LRY --with-lines --time-series \<br>--output="C:\Users\artur.tarassow\Desktop\TEST.pdf" \<br>
{ set terminal pdfcairo font 'Helvetica,15' lw 3 ; \<br>set style line 1 lt 1 lc rgb 'black' ; \<br>set style line 2 lt 2 lc rgb 'green' ; \<br>set style line 3 lt 2 lc rgb 'red' ; \<br>set style line 4 lt 2 lc rgb 'green' ; \<br>
set style line 5 lt 2 lc rgb 'orange' ; \<br>set style line 6 lt 6 lc rgb 'black' ; \<br>set ylabel '' ; set xlabel 'Horizon' ; \<br>set key bottom below ; }<br>
</div>&lt;/hansl&gt;<br><br>
</div>Artur<br>
</div></div>
valentina colombo | 9 Jul 11:31 2014
Picon

n bootstrapped coefficients

Dear Gretl's Users,


I need your help! I have to compute the confidence intervals of cumulative fiscal multipliers for horizon h=4, 8, 12, given as the ratio between GDP coefficients over Tax coefficients. I'm estimating a SVAR. The bootdata, in the model bundle, is a matrix with the bootstrap coefficient estimates (mean and median). However, to construct the confidence intervals (percentiles) of the  cumulative fiscal multipliers for each h I need the entire distribution of the n bootstrapped coefficients for each horizon h=1,..20.


How I can save the matrix of the n bootstrapped coefficients?

Any suggestions?

Thanks

Valentina

<div><div dir="ltr">

<p class="MsoNormal">Dear Gretl's Users,</p>
<p class="MsoNormal"><br></p>

<p class="MsoNormal"><span lang="EN-US">I need your
help! I have to compute the confidence intervals of cumulative fiscal
multipliers for horizon h=4, 8, 12, given as the ratio between GDP coefficients
over Tax coefficients. I'm estimating a SVAR. The bootdata, in the model
bundle, is a matrix with the bootstrap coefficient estimates (mean and median).
However, to construct the confidence intervals (percentiles) of the <span>&nbsp;</span>cumulative fiscal multipliers for each h I
need the entire distribution of the n bootstrapped coefficients for each
horizon h=1,..20. <br></span></p>
<p class="MsoNormal"><span lang="EN-US"><br></span></p>

<p class="MsoNormal"><span lang="EN-US">How I can
save the matrix of the n bootstrapped coefficients? </span></p>

<p class="MsoNormal">Any suggestions? </p>

<p class="MsoNormal">Thanks</p>

<p class="MsoNormal">Valentina</p>

 		 	   		  </div></div>
Stefano Fachin | 9 Jul 11:24 2014
Picon

matrices in the icon windows

I was trying to get a global idea of a big matrix in the matrix viewer of the icon windows and realised that the task would be much easier if I could control the number of decimal digits. Now there are many more than I need, and allowing for fewer many more columns would fit in the screen. Of course it's not a big deal :-)
as usual, thanks a lot to Allin and Jack!
Stefano
-- _________________________________________________________________________ Stefano Fachin Professore Ordinario di Statistica Economica Dip. di Scienze Statistiche Università di Roma "La Sapienza" P.le A. Moro 5 - 00185 Roma - Italia Tel. +39-06-49910834 fax +39-06-49910072 URL http://w3.uniroma1.it/fachin/
<div>
    I was trying to get a global idea of a big matrix in the matrix
      viewer of the icon windows and realised that the task
      would be much easier if I could control the number of decimal
      digits. Now there are many more than I need, and allowing for
      fewer many more columns would fit in the screen. Of
      course it's not a big deal :-)<br>
      as usual, thanks a lot to Allin and Jack!<br>
      Stefano<br>
    -- 
_________________________________________________________________________

Stefano Fachin
Professore Ordinario di Statistica Economica
Dip. di Scienze Statistiche
Universit&agrave; di Roma "La Sapienza"
P.le A. Moro 5 - 00185 Roma - Italia
Tel. +39-06-49910834
fax  +39-06-49910072
URL <a class="moz-txt-link-freetext" href="http://w3.uniroma1.it/fachin/">http://w3.uniroma1.it/fachin/</a>

  </div>
valentina colombo | 7 Jul 13:07 2014
Picon

bootstrap coefficient estimates

Dear Gretl's Users,
I have to compute the confidence intervals of cumulative fiscal multipliers for horizon=4, 8, 12, given as the ratio between GDP coefficients over Tax coefficients. I'm estimating an SVAR. The bootdata, in the model bundle, is a matrix with the bootstrap coefficient estimates (mean and median). However, to construct the confidence intervals (percentiles) I need the entire distribution of the n bootstrapped coefficients for each horizon h=1,..20

Any suggestions?

Thanks
Valentina
<div><div dir="ltr">Dear Gretl's Users,<br>I have to compute the confidence intervals of cumulative fiscal multipliers for horizon=4, 8, 12, given as the ratio between GDP coefficients over Tax coefficients. I'm estimating an SVAR. The bootdata, in the model bundle, is a matrix with the bootstrap coefficient estimates (mean and median). However, to construct the confidence intervals (percentiles) I need the entire distribution of the n bootstrapped coefficients for each horizon h=1,..20<br><br>Any suggestions? <br><br>Thanks<br>Valentina<br>
</div></div>
Picon

Testing after ARIMA or AR estimation before non linear modelling

Dear list:

First of all, sorry if the question is very basic. But It's a curiosity.

I'm exploring the issue of modelling non-linear time series. I have read in several articles that a correct strategy is often to start by fitting a linear model (for example a simple Autoregressive model) and if it is not satisfactory, then try to fit a TAR, SETAR or Markov Switching Model,.... I think this is a classical approach.

In the papers that I have read, in order to detect deviation from linearity most authors apply several test over the residuals of the simple first estimated AR model (for example RESET test, BDS test, Mc. Leod test are common in this context). Then after reject the Null in these test, they proceed with the non-linear model.

My question is: before to try with my real series, playing with some series in Gretl, I have seen that after estimating an AR model or ARIMA model with the options built in Gretl-GUI, the "test" post-estimation option only allows to test for Normality or ARCH (I think this is Mc Leod test), but other options as RESET or Non-linearity are not activated.

Only after estimating a model via OLS menu with the dependent variable and its lags -that is not the same of estimating the AR(1) model, so the constant change as usual- these post-estimation options are allowed.(I think this kind of models is the preferred process in most papers)

I think this Gretl behavior is due that in an ARIMA context you can add AR or MA terms in your model until to get residuals to be white noise. So, no more steps are needed.

However, I have checked that in Eviews the options RESET, etc. are (like not in Gretl) allowed after estimating both, the AR(1) model and the equation with the lagged dependent variable.

Y C AR(1)  AR residual term model
Y C Y(-1) model with the lagged endogenous variable

So, my curiosity and second question: Why this difference among Gretl and Eviews, the disallowed options in Gretl are for some special considerations?

With the specification of lagged dependent variable there is not poblem because the options are allowed in both software. But, if I want to perform this kind of analysis in Gretl, with an AR model, which is the correct form to proceed?. Save the residuals of my estimated AR model and export them to R or other software to perform the BDS or RESET test as usual in literature?

Thanks in advance.

José Perles
University of Alicante
Spain

 
 

I
<div><div dir="ltr">
<div>
<div>
<div>Dear list:<br><br>
</div>First of all, sorry if the question is very basic. But It's a curiosity. <br><br>
</div>I'm exploring the issue of modelling non-linear time series. I have read in several articles that a correct strategy is often to start by fitting a linear model (for example a simple Autoregressive model) and if it is not satisfactory, then try to fit a TAR, SETAR or Markov Switching Model,.... I think this is a classical approach.<br><br>
</div>
<div>In the papers that I have read, in order to detect deviation from linearity most authors apply several test over the residuals of the simple first estimated AR model (for example RESET test, BDS test, Mc. Leod test are common in this context). Then after reject the Null in these test, they proceed with the non-linear model. <br><br>
</div>
<div>My question is: before to try with my real series, playing with some series in Gretl, I have seen that after estimating an AR model or ARIMA model with the options built in Gretl-GUI, the "test" post-estimation option only allows to test for Normality or ARCH (I think this is Mc Leod test), but other options as RESET or Non-linearity are not activated. <br><br>
</div>
<div>Only after estimating a model via OLS menu with the dependent variable and its lags -that is not the same of estimating the AR(1) model, so the constant change as usual- these post-estimation options are allowed.(I think this kind of models is the preferred process in most papers)<br><br>I think this Gretl behavior is due that in an ARIMA context you can add AR or MA terms in your model until to get residuals to be white noise. So, no more steps are needed.<br>
</div>
<div><br></div>
<div>However, I have checked that in Eviews the options RESET, etc. are (like not in Gretl) allowed after estimating both, the AR(1) model and the equation with the lagged dependent variable.<br><br>
</div>
<div>Y C AR(1)&nbsp; AR residual term model <br>
</div>
<div>Y C Y(-1) model with the lagged endogenous variable <br>
</div>
<div><br></div>
<div>So, my curiosity and second question: Why this difference among Gretl and Eviews, the disallowed options in Gretl are for some special considerations?<br>
</div>
<div><br></div>
<div>With the specification of lagged dependent variable there is not poblem because the options are allowed in both software. But, if I want to perform this kind of analysis in Gretl, with an AR model, which is the correct form to proceed?. Save the residuals of my estimated AR model and export them to R or other software to perform the BDS or RESET test as usual in literature?<br>
</div>
<div><br></div>
<div>Thanks in advance.<br><br>
</div>
<div>Jos&eacute; Perles<br>
</div>
<div>University of Alicante<br>
</div>
<div>Spain<br>
</div>
<div>
<br>&nbsp;<br>
</div>
<div>&nbsp;<br>
</div>
<div><br></div>I <br>
</div></div>
Logan Kelly | 3 Jul 21:43 2014

Test for data connection

Hello all,

 

Is there a way to test for an active data connection before calling readfile() or curl()?

 

Thanks,

 

Logan

<div>
<div class="WordSection1">
<p class="MsoNormal">Hello all,<p></p></p>
<p class="MsoNormal"><p>&nbsp;</p></p>
<p class="MsoNormal">Is there a way to test for an active data connection before calling readfile() or curl()?<p></p></p>
<p class="MsoNormal"><p>&nbsp;</p></p>
<p class="MsoNormal">Thanks,<p></p></p>
<p class="MsoNormal"><p>&nbsp;</p></p>
<p class="MsoNormal">Logan<p></p></p>
</div>
</div>

Gmane