ARIMA model with exogenous variable

Dear GRTEL users, 

I have estimated the follwing models

1) ARIMAX of order (2,1,0) with additional constant for the dependent variable GMSL_CW (endogenous) and
the independent variable GISS_GSST (exogenous) (arima 2 1 0; GMSL_CW const GISS_GSST)

Function evaluations: 24
Evaluations of gradient: 8

Model 4: ARMAX, using observations 1881-2001 (T = 121)
Estimated using Kalman filter (exact ML)
Dependent variable: (1-L) GMSL_CW
Standard errors based on Hessian

              coefficient   std. error   t-ratio   p-value 
  ---------------------------------------------------------
  const         1.69314     0.281737      6.010    1.86e-09 ***
  phi_1        -0.361518    0.0878460    -4.115    3.87e-05 ***
  phi_2        -0.250534    0.0898182    -2.789    0.0053   ***
  GSST_GISS     3.72863     1.36854       2.725    0.0064   ***

Mean dependent var   1.497521   S.D. dependent var   5.290911
Mean of innovations -0.015051   S.D. of innovations  4.819733
Log-likelihood      -362.0990   Akaike criterion     734.1979
Schwarz criterion    748.1769   Hannan-Quinn         739.8753

                        Real  Imaginary    Modulus  Frequency
  -----------------------------------------------------------
  AR

Re: Edit attributes dialog box

I also tested the new functionality and it really works nicely
enhancing usability. May I suggest:
1)- Making the sort on click feature a little more subtle? I think
highlighting the column heads and changing the color of the column can
be a bit distracting. Just the little triangle is enough no?
2)- For switching variables in the attributes button, is it possible
to use arrow signs instead? This would also deemphasized the two new
icons a little.
Cheers
Talha

On Tue, Jun 30, 2009 at 11:08 PM, Allin Cottrell<cottrell@...> wrote:
> On Thu, 25 Jun 2009, Talha Yalta wrote:
>
>> I think it would be really neat if there was a left and right
>> arrow next to the variable name in the "edit attibutes" window
>> so that one can easily edit or update a bunch of variables,
>> their descriptions etc. without closing an reopening the window
>> each time. What do you think?
>
> Good idea, and it's now implemented -- well, via Up and Down
> buttons.  You can also select the variable to edit, with the Edit
> Attributes dialog open, by clicking on its row in the main window.
> Changing the variable in this way does not automatically commit
> any changes to the current variable: use the "Apply" button to do
> that first.
>
> Also, responding to an older request, the column headings in the
> gretl main window are now clickable and can be used to sort the
> variables by ID number or name, ascending or descending.

Re: Edit attributes dialog box

On Wed, 1 Jul 2009, Talha Yalta wrote:

> I also tested the new functionality and it really works nicely
> enhancing usability. May I suggest:
> 1)- Making the sort on click feature a little more subtle? I think
> highlighting the column heads and changing the color of the column can
> be a bit distracting. Just the little triangle is enough no?

That's all controlled by GTK and its theme/style files.  But I
have now quieted things a little: when you revert to the default
setting for the main window (variables sorted by ascending ID
number), the extra visual stuff associated with sorting
disappears.

> 2)- For switching variables in the attributes button, is it possible
> to use arrow signs instead? This would also deemphasized the two new
> icons a little.

I have made these buttons more discreet.  But the actual image you
see depends on the GTK theme in force; I'm just using
GTK_STOCK_GO_UP and GTK_STOCK_GO_DOWN, which with the standard GTK
theme produce arrows.

Allin.

Re: ARIMA model with exogenous variable

On Wed, 1 Jul 2009 Didier.Monselesan@... wrote:

> I have estimated the follwing models
>
> arima 2 1 0; GMSL_CW const GISS_GSST
> arima 2 0 0; d_GMSL_CW const GISS_GSST
>
> As expected the estimated model parameters are identical in both
> cases...  However, I could not explain how GRETL computes the
> fitted values and residuals for the levels (not the differences)
> in model 1). I was expecting the fitted values of model 1) to be
> the cumulated sums of fitted values of model 2)...

I'm attaching a script giving a version of the demonstration in
the Gretl User's Guide (Example 20.1 on ARIMA forecasting). I have
modified the example to resemble more closely the model you're
estimating, and have added a comparison of fitted values (as
accessed via $yhat) alongside the comparison of results from
the "fcast" command.

Hopefully this should show you what's going on.

Allin Cottrell

open greene18_2.gdt
# log of quarterly U.S. nominal GNP, 1950:1 to 1983:4
genr y = log(Y)
# and its first difference
genr dy = diff(y)
# log of real money supply (for ARMAX)

Re: Edit attributes dialog box

> I have made these buttons more discreet.  But the actual image you
> see depends on the GTK theme in force; I'm just using
> GTK_STOCK_GO_UP and GTK_STOCK_GO_DOWN, which with the standard GTK
> theme produce arrows.
Thanks. I now see up and down arrows even though I didn't change the
GTK theme. It looks nice but my only complaint is that the arrows are
green and these are the only colored icons in the whole GUI. If this
is not something easy to control, how about just using "<-" and "->"
as in various menus such as the model specification menu?

Also, the new sorting functionality is great but I think it needs to
include the "Descriptive label" column as well for the sake of
consistency and completeness.
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)
--

Re: Edit attributes dialog box

On Mon, 6 Jul 2009, Talha Yalta wrote:

> > I have made these buttons more discreet.  But the actual image you
> > see depends on the GTK theme in force; I'm just using
> > GTK_STOCK_GO_UP and GTK_STOCK_GO_DOWN, which with the standard GTK
> > theme produce arrows.
> Thanks. I now see up and down arrows even though I didn't change the
> GTK theme. It looks nice but my only complaint is that the arrows are
> green and these are the only colored icons in the whole GUI.

You must be using an odd GTK theme if that's the case .

> Also, the new sorting functionality is great but I think it needs to
> include the "Descriptive label" column as well for the sake of
> consistency and completeness.

OK, done.

Allin.

Weighted data?

Hello all, 

I am a fairly new user of Gretl, and I like it very much this far. One
problem I have is that I am currently working on survey data and it
seems that Gretl does not have built-in commands to assign weights to
the data, is this correct? Thanks for any help,

Andrea Giusto

Re: Weighted data?

On Tue, 7 Jul 2009, Andrea Giusto wrote:

> I am a fairly new user of Gretl, and I like it very much this
> far.

Welcome to the gretl list!

> One problem I have is that I am currently working on survey data
> and it seems that Gretl does not have built-in commands to
> assign weights to the data, is this correct? Thanks for any
> help,

Yes, you're right, gretl does not have built-in handling for
survey data with observation weights at present.  Of course, we do
offer weighted least squares, but the user has to supply the
weights series.  This is something we should probably consider.

Allin Cottrell

reversing / flipping a matrix

Dear gretl users,

yesterday I needed to flip a matrix/vector around by reversing the order
of the elements. (I guess like rev() or flipud() in other languages.)
But I didn't find the function to do that, so I used an inefficient
workaround using msortby(). But since this seems like quite basic
functionality I think I either missed something or it should be added,
right?

thanks,
sven

Re: reversing / flipping a matrix

On Fri, 10 Jul 2009, Sven Schreiber wrote:

> Dear gretl users,
>
> yesterday I needed to flip a matrix/vector around by reversing the order
> of the elements. (I guess like rev() or flipud() in other languages.)
> But I didn't find the function to do that, so I used an inefficient
> workaround using msortby(). But since this seems like quite basic
> functionality I think I either missed something or it should be added,
> right?

Right. I must have said to myself "I need to do this. Oh well, not now" a 
thousand times. It's now in CVS. The user-level function is called rev(). 
Example:

? A = rev(I(3))
  A = rev(I(3))
Generated matrix A
? print A
  print A
A (3 x 3)

   0   0   1
   0   1   0
   1   0   0

Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche