plot saving error
by Stefano
The steps followed are pretty trivial: I run the code pasted below,
looked at the plot in its window, right-clicked to open the menu and
asked to save the plot as EPS. Now I cannot reproduce it myself, so I
guess it must have been some sort of mess in the memory management
(using Windows 10 that would not be surprising at all). In fact that's
what I thought rightaway, I posted the question mostly in case it may be
of interest for developers. Thanks for your interest!
Stefano
<hansl>
plot
options with-lines
literal set nokey
literal set xrange [15:85]
literal FRF(x) = 7.984 -0.522*x + 0.011*x**2 -0.00007*x**3
printf "plot FRF(x)"
end plot --output=display
</hansl>
--
________________________________________________________________________
Stefano Fachin
Professore Ordinario di Statistica Economica
Dip. di Scienze Statistiche
"Sapienza" Università di Roma
P.le A. Moro 5 - 00185 Roma - Italia
Tel. +39-06-49910834
fax +39-06-49910072
web http://stefanofachin.site.uniroma1.it/
--
________________________________________________________
Le informazioni
contenute in questo messaggio di posta elettronica sono strettamente
riservate e indirizzate esclusivamente al destinatario. Si prega di non
leggere, fare copia, inoltrare a terzi o conservare tale messaggio se non
si è il legittimo destinatario dello stesso. Qualora tale messaggio sia
stato ricevuto per errore, si prega di restituirlo al mittente e di
cancellarlo permanentemente dal proprio computer.
The information contained
in this e mail message is strictly confidential and intended for the use of
the addressee only. If you are not the intended recipient, please do not
read, copy, forward or store it on your computer. If you have received the
message in error, please forward it back to the sender and delete it
permanently from your computer system.
--
2 years, 11 months
plot saving error
by Stefano
Gretl 2020e, build date 2020-11-21, Gnuplot as bundled with Gretl, OS
Windows 10. Some time after posting the message I tried again and it
worked all right, so the error
("C:\Users\pc\AppData\Roaming\gretl\.gretl-8664\graph.1: No such file or
directory") appears to be random, which in some respects is even worse...
thanks
Stefano
--
________________________________________________________________________
Stefano Fachin
Professore Ordinario di Statistica Economica
Dip. di Scienze Statistiche
"Sapienza" Università di Roma
P.le A. Moro 5 - 00185 Roma - Italia
Tel. +39-06-49910834
fax +39-06-49910072
web http://stefanofachin.site.uniroma1.it/
--
________________________________________________________
Le informazioni
contenute in questo messaggio di posta elettronica sono strettamente
riservate e indirizzate esclusivamente al destinatario. Si prega di non
leggere, fare copia, inoltrare a terzi o conservare tale messaggio se non
si è il legittimo destinatario dello stesso. Qualora tale messaggio sia
stato ricevuto per errore, si prega di restituirlo al mittente e di
cancellarlo permanentemente dal proprio computer.
The information contained
in this e mail message is strictly confidential and intended for the use of
the addressee only. If you are not the intended recipient, please do not
read, copy, forward or store it on your computer. If you have received the
message in error, please forward it back to the sender and delete it
permanently from your computer system.
--
2 years, 11 months
plot saving error
by Stefano
dear all,
after asking to save a plot as EPS file in a personalised directory I
got the following error message:
"C:\Users\pc\AppData\Roaming\gretl\.gretl-8664\graph.1: No such file or
directory". "Copy to clipboard" "display PDF" produce the same result.
Any hints?
thanks,
Stefano
--
________________________________________________________________________
Stefano Fachin
Professore Ordinario di Statistica Economica
Dip. di Scienze Statistiche
"Sapienza" Università di Roma
P.le A. Moro 5 - 00185 Roma - Italia
Tel. +39-06-49910834
fax +39-06-49910072
web http://stefanofachin.site.uniroma1.it/
--
________________________________________________________
Le informazioni
contenute in questo messaggio di posta elettronica sono strettamente
riservate e indirizzate esclusivamente al destinatario. Si prega di non
leggere, fare copia, inoltrare a terzi o conservare tale messaggio se non
si è il legittimo destinatario dello stesso. Qualora tale messaggio sia
stato ricevuto per errore, si prega di restituirlo al mittente e di
cancellarlo permanentemente dal proprio computer.
The information contained
in this e mail message is strictly confidential and intended for the use of
the addressee only. If you are not the intended recipient, please do not
read, copy, forward or store it on your computer. If you have received the
message in error, please forward it back to the sender and delete it
permanently from your computer system.
--
2 years, 11 months
vec() and vech(): row-major vs. column-major
by Sven Schreiber
Hi,
I just found out that gretl's vec() and vech() functions have an
important difference: As duly documented, vec() stacks the columns (as
expected), but a little bit to my surprise vech() stacks the (relevant
parts of the) rows of the input. The latter property is not very
explicitly documented. (It just talks about the upper triangle of the
matrix, which at least is a hint in that direction.)
So basically to get the Wikipedia-style convention of the vech function
(https://en.wikipedia.org/wiki/Vectorization_(mathematics)#Half-vectorization)
we need to work on the transpose, vech(m').
This message just serves to share this "discovery". I'm not saying we
should change vech's behavior because that would break too much stuff.
But maybe a more explicit note should be added to the documentation.
(You may wonder why this matters at all, given that the input typically
should be symmetric; but I had to work with some code that just fills
the lower triangle of a matrix and then expands this to a symmetric
matrix. So the input there wasn't symmetric, by design.)
thanks
sven
2 years, 11 months
operation on arrays
by Stefano
This is entirely trivial, but anyway: I realised that the manual somehow
underplays the possible operations on arrays ("At present only one
operation is available for arrays as a whole, namely appending", p. 78),
as subsets of an array can be extracted using matrix syntax, as in
<hansl>
strings S = array(3)
S[1] = "a"
S[2] = "b"
S[3] = "c"
strings Q = S[1:2]
</hansl>
which starting from the array "a","b","c" produces the array "a","b".
This actually came handy to me (I started with a panel of units and
their labels in an array, then dropped some units and needed to redefine
the array), so it may be worth it mentioning it in the manual.
bye
Stefano
________________________________________________________________________
Stefano Fachin
Professore Ordinario di Statistica Economica
Dip. di Scienze Statistiche
"Sapienza" Università di Roma
P.le A. Moro 5 - 00185 Roma - Italia
Tel. +39-06-49910834
fax +39-06-49910072
web http://stefanofachin.site.uniroma1.it/
--
________________________________________________________
Le informazioni
contenute in questo messaggio di posta elettronica sono strettamente
riservate e indirizzate esclusivamente al destinatario. Si prega di non
leggere, fare copia, inoltrare a terzi o conservare tale messaggio se non
si è il legittimo destinatario dello stesso. Qualora tale messaggio sia
stato ricevuto per errore, si prega di restituirlo al mittente e di
cancellarlo permanentemente dal proprio computer.
The information contained
in this e mail message is strictly confidential and intended for the use of
the addressee only. If you are not the intended recipient, please do not
read, copy, forward or store it on your computer. If you have received the
message in error, please forward it back to the sender and delete it
permanently from your computer system.
--
2 years, 11 months
Authoring a contributed function package: your winter corona break project?
by Sven Schreiber
Hi everybody,
we have close to 150 contributed function packages that can be
downloaded from within gretl. For the list see either on the web
http://ricardo.ecn.wfu.edu/gretl/cgi-bin/gretldata.cgi?opt=SHOW_FUNCS,
or from the gretl menus choose File/Function packages/On server...
For those who don't know it yet, such packages are written in the Hansl
scripting language, which is quite easy and intuitive but has become
very powerful over the years. It mainly supports a functional
programming paradigm, it is not object-oriented for example. For more
information about the language see the "Hansl primer" PDF document under
gretl's Help menu.
Now if you're sitting in a place of the world where the pandemic is not
under control (for example Europe or the Americas) and are getting bored
from all this working-from-home and everything being closed, why not
implement some missing econometric feature in the form of such a
function package? Of course there's a learning curve involved, but in
principle it is not very difficult. We have an official Gretl Function
Package Guide also in PDF format under the Help menu. And last year
Stefano contributed another very nice step-by-step tutorial:
http://gretl.sourceforge.net/gfnguide/gfn_for_dummies.html
(If you know of more resources like that one, please share them here!)
For any further questions just use this mailing list, especially if the
answer is likely to be of general interest. Apart from that you could
also ask some packaging-related questions to the team of package
moderators if you prefer to not discuss it in public; apart from the
gretl heroes Allin and Jack the moderation team consists of Artur
Tarassow and myself. (Please grab our email addresses from previous
mailing list postings, I don't want to repeat them here inline. And
please only function packaging-related questions off-list, no other
general gretl topics.)
As a final remark, for a first function package I would suggest to keep
it simple and not aim for something overly complex. A package can always
be upgraded later (they also have version numbers). An example of a very
small function package is "criteria" from Allin himself, which basically
just has 18 lines of code. But a package also shouldn't get much smaller
than that :-)
cheers
sven
2 years, 12 months
split dependent variable
by Stefano
The problem reminds me of the issue "forecasting the aggregate vs
summing disaggregate forecasts". Hendry and Hubrich (2011) "Combining
Disaggregate Forecasts or Combining Disaggregate Information to Forecast
an Aggregate",Journal of Business & Economic Statistics, conclude on the
basis of the RMSFEs (no tests) that direct forecasting of the aggregate
is better than combining disaggregate forecasts. In this persepctive you
could use a Diebold-Mariano test.
Stefano
--
________________________________________________________________________
Stefano Fachin
Professore Ordinario di Statistica Economica
Dip. di Scienze Statistiche
"Sapienza" Università di Roma
P.le A. Moro 5 - 00185 Roma - Italia
Tel. +39-06-49910834
fax +39-06-49910072
web http://stefanofachin.site.uniroma1.it/
--
________________________________________________________
Le informazioni
contenute in questo messaggio di posta elettronica sono strettamente
riservate e indirizzate esclusivamente al destinatario. Si prega di non
leggere, fare copia, inoltrare a terzi o conservare tale messaggio se non
si è il legittimo destinatario dello stesso. Qualora tale messaggio sia
stato ricevuto per errore, si prega di restituirlo al mittente e di
cancellarlo permanentemente dal proprio computer.
The information contained
in this e mail message is strictly confidential and intended for the use of
the addressee only. If you are not the intended recipient, please do not
read, copy, forward or store it on your computer. If you have received the
message in error, please forward it back to the sender and delete it
permanently from your computer system.
--
3 years
Re: split dependent variable: looking for a test
by Sven Schreiber
Am 01.12.2020 um 16:09 schrieb Summers, Peter:
> Hi Allin,
>
> Your logic sounds similar to that of the PE Test (MacKinnon, White & Davidson, 1983, J of Econometrics) for comparing models with y vs log(y). I hadn't heard about it until a couple weeks ago while I was teaching about LM tests. Here's a link to the R petest() function: https://www.rdocumentation.org/packages/lmtest/versions/0.9-38/topics/petest.
Hi, such a test is also available in gretl as part of the contributed
package BoxCoxFuncForm, me being a coauthor there. The reference we have
included in the helpt text is:
Testing Linear and Loglinear Regressions against Box-Cox Alternatives
Author(s): Russell Davidson and James G. MacKinnon
Source: The Canadian Journal of Economics / Revue canadienne
d'Economique,
Vol. 18, No. 3 (Aug., 1985), pp. 499-517
Stable URL: https://www.jstor.org/stable/135016
However, I'm not sure if that is what Allin was asking for.
cheers
sven
3 years
split dependent variable: looking for a test
by Allin Cottrell
This question is not gretl-specific, but answers might show up gretl
features and hopefully it may be of interest.
My dependent variable is the quantity demanded of a certain good, an
hourly time series extending over two years. It exhibits strong
seasonality, both by hour of the day and season of the year; it has
an upward trend; and it's clearly affected by various measures of
weather (temperature, humidity, wind speed). Plain OLS produces
quite a decent fit, but by inspecting the loglikelihood-for-level
figure produced by gretl (via the Jacobian) I can see that taking
the log of the dependent variable gives a better fit. All fine.
However, total demand is the sum of demand from two classes of
consumer -- call them A and B -- and I'm wondering if a better fit
can be obtained by summing the fitted values from separate
regressions, with dependent variables the demand from consumers A
and B respectively. (Note: a Chow test in dummy variable mode is not
applicable, the unit of observation is the hour, not the
transaction.)
My thought was: compute two SSRs in levels, using (restricted) the
exponentiated fitted values from the overall model and
(unrestricted) the sum of the exponentiated fitted values from the
two consumer-class models, then calculate an F test based on the
difference of SSRs in the usual way.
Questions: Does this sound valid? Is there a better way of doing it?
[I'm aware of debate over how best to produce predictions of levels
from log-linear regression, but I'm not sure quite how it applies in
this case.]
Allin Cottrell
3 years
