Hello all.. following the reference manual I generated a dummy
variable for year 1992 (1992=1, 0 elsewhere) with command:
genr tpD92 = (obs=1992)
The dummy is created without problems and I can see it.. however when
I try to use the new serie in estimations (hsk) I receive an error
"Not numerical value in a calculation".. why?? Is "1" treated as
integer instead of a numerical value?? If that is the case, how can I
convert the dummy serie in a "normal" serie ?
Up ot now I was using exogenously-loaded dummy variables, but let them
beeing generated by gretl would be faster...
antonello a.t lobianco.org
I am working on a problem where I want to backcast a time series, i.e.,
forecast into the past to a time before the measurements of the times
series were available.
None of the software packages I have access to (including Gretl) appear
to offer an explicit backcasting option. I have read that a stationary
time series is time-invariant, i.e., it has the same properties forwards
and backwards in time. So, when I go to backcast the data is it valid to
analyze the data in Gretl in reverse order and just do a regular forecast?
I'm sorry the following questions are going to be really basic but I've got
to start somewhere...
I have a time series data (temperature against time for example) and I need
to predict what the temperature is going to be in the future. Is this the
sort of thing that Gretl will do for me? I've had a play with Gretl and I've
managed to get it to give me a couple of graphs and I've even managed to
perrform a forecast run but the forecasts I'm getting are always straight
lines when I would have expected something that looks like the data I have.
I think what I need to learn about is statistics in general and regression
analysis in particular but I'm not a mathematician / statistician so I don't
know if that is correct (I'm a chemist turned programmer). I'm willing to
learn what I need to know but right now I a bit stuck just figuring out what
it is I should learn.
Many thanks for any pointers, Graham Smith
cottrell(a)wfu.edu @ INTERNET skrev 2008-04-03 00:26:33 :
> On Wed, 2 Apr 2008, Thomas La Bone wrote:
> > Hunting through the Gretl menus and documentation I can find no >
> mention of the Holt or Holt-Winters forecasting methods. Will >
> Gretl perform Holt-Winters forecasting?
> Yes, if you write a script to do it. That is, it's well within
> gretl's capabilities but there's no built-in Holt-Winters command.
> Alin Cottrell
Thomas, you can write a feature request for this feature here:
IMHO, it would be a useful feature to include in gretl.
I've just downloaded and installed gretl, but the graphs don't work. I've
uninstalled and reinstalled but it still doesn't work. When I try to get a
graph, it opens a gnuplot window and gives an error message: "Invalid byte
sequence in conversion input". In fact the message is "sequência de bytes
inválida na entrada de conversão" since I am runnig a Portuguese-Brazilian
Well, I also noticed that gretl uses a directory in "Documents and
Settings/Application Data" for temp files. When I ask for any graph, it
generates the .PNG temp file and gives the error above. When I click the
"ok" button, the png and other temp files are deleted.
My apologies - the URL is:
(3 'w' - instead of 4)
On 02/04/2008, Philipp K. Janert <janert(a)mailaps.org> wrote:
> Finally: A book on Gnuplot!
> If you are interested in data analysis and visualization,
> you might be interested in the upcoming book:
> "Gnuplot in Action"
> (to be published, Fall 2008, Manning Publications).
> You can pre-order it directly from the publisher:
> Gnuplot is probably the most popular open-source
> program for plotting and visualization. The book
> assumes no previous familiarity with Gnuplot. It
> introduces the most basic concepts rapidly, and
> then moves on to explain Gnuplot's advanced
> concepts and power features in detail.
> If you want to learn more about the book and the
> author, check out my book page at
> If you have comments or suggestions for the book,
> please let me know!
> Feel free to forward this information as appropriate.
> Help-octave mailing list
Finally: A book on Gnuplot!
If you are interested in data analysis and visualization,
you might be interested in the upcoming book:
"Gnuplot in Action"
(to be published, Fall 2008, Manning Publications).
You can pre-order it directly from the publisher:
Gnuplot is probably the most popular open-source
program for plotting and visualization. The book
assumes no previous familiarity with Gnuplot. It
introduces the most basic concepts rapidly, and
then moves on to explain Gnuplot's advanced
concepts and power features in detail.
If you want to learn more about the book and the
author, check out my book page at
If you have comments or suggestions for the book,
please let me know!
Feel free to forward this information as appropriate.