Re: [Gretl-users] Ox support on OS/X
by Henrique Andrade
Dear Sven,
I'll try this. Thanks for your help!
Best,
Henrique
Enviado via Palm
-----Original Message-----
From: Sven Schreiber <svetosch(a)gmx.net>
Subj: Re: [Gretl-users] Ox support on OS/X
Date: Fri 4 Sep 2009 7:09
Size: 513 bytes
To: Gretl list <gretl-users(a)lists.wfu.edu>
Am 04.09.2009 05:07, Henrique schrieb:
> ? foreign language=Ox
> #include <oxstd.h>
> #import <packages/dpd/dpd>
> /Users/henrique/.gretl/gretltmp.ox (3): 'packages/dpd/dpd.h' include
> file not found
>
>
> What can I do to fix this error?
>
tell ox where dpd.h is? (because apparently it's not finding it in the
specified path)
-sven
_______________________________________________
Gretl-users mailing list
Gretl-users(a)lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users
17 years
saving Forecasts
by Wolfgang Polasek
After ARIMA:
Is there a trick to save the forecast e.g. from 2010 to 2015 after the
observed data until 2009?Thanks Wolf
17 years
problems with gretl libr foreign blocks under win32
by chris
Hi,
First of all it seems that currently the default setting is to use libr
and not rterm.exe if avaiable. I appreciate that a lot! :)
But there are two problems:
On my system the "Rsrc" file created for R contains two sink(<file>)
commands at the beginning of the file and just one sink() command at the
end. So after several foreign blocks, R fails with the error message
"sink stack is full". The error message appears only when using the
gretl command client. gretlw32 simply stops executing the script saying
"external command failed".
A second problem is that sometimes R complaints that it cannot find its
dlls. This seems to be well known to other users of libr already, see:
http://code.env.duke.edu/projects/mget/ticket/203.
I added the R-bin directory to my path environment variable and now the
problem is gone. I don't know whether we should add a workaround for
this problem, but if not we should at least give a hint in a helpfile
somewhere.
chris
17 years
ARIMA on electricity market prices
by Oliver Heering
Hi (again),
i am about to estimate an ARIMA model to hourly electricity price
data. The data has cyclic characteristics at 24 hours and 168 hours (a
week). I actually found this paper
(http://www.uclm.es/area/gsee/JavierC/papers/01216141.pdf) where an
ARIMA model was fitted to about the same electricity price data (same
characteristics), but i am unable to transfer their results to gretl.
Can anyone help me finding the right AR/MA/Differencing and seasonal
orders for this data? I already loaded the data into gretl as hourly
timeseries data, but i have no clue about choosing the right ARIMA
orders to get anything remotely close to a good fit. The paper
mentions the numbers 24 and 168 in their model but i can't enter
anything above 10 as AR/MA orders and the seasonal orders are limited
to 4 as well (assuming quarterly data?). But as i said, i'm not even
sure if i need to enter numbers this big as the paper also mentions
that the models only use the previous 5, respective 2 hours.
Any questions regarding the data can be answered by looking at the
above paper. My data is about exactly the same.
Thanks.
--
Oliver
17 years
GRETL string concatenation for use in SQL statement
by Mike Pfeiff
I am very new to GRETL so all help is greatly appreciated.
Previously I have asked about making an ODBC connection to MS Access to
bring in my own data. With the help of users in this community I have
been successful in that end ever.
Below is the code that I used:
nulldata 250
setobs 12 1990:1
open dsn=FCST_INPUTS --odbc
string sqlstr="SELECT Emplopy FROM Data_Table_to_GRETL WHERE
census_division="East North Central"
data Employ @sqlstr --odbc
setinfo MWh -n "in MWh"
Now I have a simple string concatenation question to be used in a SQL
statement.
I need to bring in two (2) data series from the same MS Access table
named Data_Table_to_GRETL. The two series are: (1) Employ; and (2) Pop
Both of data series need to be from the same census_division (in the
Acess database I have data by year month for each of the 9 US census
divisions). In this example the census division is "East North
Central".
I though that I could define a string named census_divicsion early in
the code and then concatenate the sqlstr for each "ODBC read" as
follows:
string census_division="East North Central"
nulldata 250
setobs 12 1990:1
# bring in Employment data
open dsn=FCST_INPUTS --odbc
string sqlstr="SELECT Employ FROM Data_Table_to_GRETL WHERE
census_division=" ~ census_division
data Employ @sqlstr --odbc
# bring in Population data
open dsn=FCST_INPUTS -odbc
string sqlstr2="SELECT Pop FROM Data_Table_to_GRETL WHERE
census_division="~ census_division
data Pop sqlstr2 --odbc
However, for some reason even thought the string appears to be correct,
it errors out:
Error executing script: halting
>data Employ @sqlstr -odbc
Any assistance providing the correct logic for concatenating the sqlstr
that would make it able to be read in SQL would be greatly appreciated.
Sincerely,
Mike
17 years
Applying an ARIMA Model
by Oliver Heering
Hi,
coming from data mining and machine learning i have the following
question which may or may not sound stupid for you, but as i have no
clue and there are no stupid questions (just stupid answers) i take my
chance posting the question to this list:
I already figured out how to fit an ARIMA process to my data (ok, how
to actually find GOOD AR/MA/difference orders would be another
question). And i can save the resulting model as Icon to my session.
Does that mean i can also apply my model to any other data (of the
same type of course)? Let's say i fit my model against a timeseries
range A. How can i find out how my model fits to another timeseries
range B?
Maybe this isn't possible at all and i am misunderstanding the whole
concept of ARIMA models and -forecasting. I am primary working with
the data mining tool "RapidMiner", which allows you to easily apply
any learned model to new and unseen data (classification mainly, but
regression as well) and i wonder if it is the same with an estimated
ARIMA model.
Thanks (again) for any answer. :-)
By the way, i appreciate that case markers can now be up to 15 chars
in length, but what are they actually used for? Of course i'd like to
have them show up on my plots, which i currently do by manually
tweaking the gnuplot commands, but i think this functionality is still
missing in gretl, am i correct? So what's left? Is there any practical
use-case for case markers in timeseries analysis?
--
Oliver
17 years
