session error
by Summers, Peter
Hi folks,
I created a session file yesterday that included some graphs that I'd been editing. Now though, I can't re-open the session. Gretl says "xmlParseFile failed on .graphs of components\session.xml", and I can open other sessions with no problem. So I'm guessing this has to do with the edited plot commands? I've attached a copy if that helps.
TIA!
PS
===============================
Dr. Peter Summers
Assistant Professor
Department of Economics
Texas Tech University
===============================
15 years, 1 month
VAR Forecast
by Henrique Andrade
Dear Gretl Community,
I'm trying to perform forecasts using a VAR model. Please take a look
at the following script:
<script>
open australia.gdt
"VAR 1" <- var 1 E PAU PUS
fcast 1991:2 1991:4 --out-of-sample
matrix Previsao = $fcast
matrix ErroPadrao = $fcerr
</script>
The VAR estimation goes just fine, but the forecasting exercise gives
me this error warning:
<output>
? fcast 1991:2 1991:4 --out-of-sample
Data error
Error executing script: halting
> fcast 1991:2 1991:4 --out-of-sample
</output>
What I doing wrong?
Best,
--
Henrique C. de Andrade
Doutorando em Economia Aplicada
Universidade Federal do Rio Grande do Sul
www.ufrgs.br/ppge
15 years, 1 month
Re: [Gretl-users] glitch with X-12
by Allin Cottrell
On Mon, 11 Oct 2010, Summers, Peter wrote:
> Allin: The latest version does seem to have resolved the
> "mini-error box" problem I was having. Now, though, there seems
> to be a minor gnuplot issue if I don't want the graphs.
Thanks for the report, That's now fixed.
Allin Cottrell
15 years, 1 month
Re: [Gretl-users] glitch with X-12
by Allin Cottrell
On Fri, 8 Oct 2010, John C Frain wrote:
> I have tried several sequences of adjustments. if x12arima warnings
> are not generated all appears to work well. If a warning is generated
> the warning box is repeated at all subsequent runs of x12arima in the
> current session. (If there is no warning a small empty box is
> displayed). I presume that a switch is being set to display the
> "warning" box but it is not being reset after display.
True enough; now fixed.
> I think that the trading day button in the interface might be better
> off by default. One needs to switch it on for flow data which show
> trading day effects.
Ok, I'll defer to you on that.
Allin Cottrell
15 years, 1 month
Can the results of forecasts be saved?
by Vivien Lee
Hello,
I'm new to gretl. So please bear with me.
I ran an ARIMA (2,1) model and then did an out-of-sample forecast. Original sample is set to use the first 1000 observations and the forecast is on observations 1001 to 1050. I want to save the predicted y values and then plot it along with the actual and fitted values and the unconditional mean. I can't seem to find any command to save the predicted values. Even if I can add the predicted values to the data set, I'm not sure how to include the unconditional mean on the graph.
Thank you in advance for your help,
vivien
15 years, 1 month
Re: [Gretl-users] glitch with X-12
by Allin Cottrell
On Fri, 8 Oct 2010, Summers, Peter wrote:
> Many thanks Allin! However, when I tried to install your latest
> update, I get a message that the setup files are corrupted.
Hmm, there were no visible errors on building the distribution
files -- maybe the upload went wrong somehow? I've now done a
new upload.
Allin
15 years, 1 month
Re: [Gretl-users] glitch with X-12
by Allin Cottrell
On Fri, 8 Oct 2010, Summers, Peter wrote:
> I just did some seasonal adjustment with the latest Windows
> snapshot (build date 10/8). With one series I got a warning
> about at least one significant trading day effect. I had a look,
> closed the warning box & adjusted the rest of my data. Some
> subsequent data also generated a warning, but the warning box
> was shrunk - just the triangle-exclamation point sign and a
> "close" button.
Probably the x12a "err" file actually said nothing but we got the
mistaken impression that it had some content. (x12a always
produces such a file, containing some "boilerplate" and blank
lines, if nothing else.) I've now adjusted the relevant test, so
hopefully you shouldn't get false positives. (CVS and snapshots
updated)
Allin Cottrell
15 years, 1 month
glitch with X-12
by Summers, Peter
Folks,
I just did some seasonal adjustment with the latest Windows snapshot (build date 10/8). With one series I got a warning about at least one significant trading day effect. I had a look, closed the warning box & adjusted the rest of my data. Some subsequent data also generated a warning, but the warning box was shrunk - just the triangle-exclamation point sign and a "close" button.
PS
===============================
Dr. Peter Summers
Assistant Professor
Department of Economics
Texas Tech University
===============================
15 years, 1 month
Re: [Gretl-users] Problems with graphs
by Allin Cottrell
On Thu, 7 Oct 2010, Henrique Andrade wrote:
> I think I've found a bug in Gretl (version 1.9.1cvs, build date 2010-10-05,
> under Windows Vista). Everytime I try to delete 2 graphs Gretl crashes. I'm
> sending a copy of a Gretl session where the problem occurs.
Thanks. This should now be fixed in CVS and shapshots.
Allin
15 years, 1 month
Re: [Gretl-users] Problems with graphs
by Allin Cottrell
On Thu, 7 Oct 2010, Henrique Andrade wrote:
> I think I've found a bug in Gretl (version 1.9.1cvs, build date 2010-10-05,
> under Windows Vista). Everytime I try to delete 2 graphs Gretl crashes. I'm
> sending a copy of a Gretl session where the problem occurs.
Confirmed. Hopefully this should be a fairly quick fix.
Allin
15 years, 1 month
