Supremum (sup) of a time series
by Gabriela Nodari
Dear all,
Is there a command to identify the supremum of a time series in gretl?
Thank you in advance...
Gabriela
11 years, 9 months
Problem with xlsx file
by Henrique Andrade
Dear Allin,
I'm trying to open an Excel spreadsheet but Gretl shows me an error
message. I don't understand what is going wrong once I've worked with this
data before with no issues (please find attached the "teste.xlsx" file).
I'm using the following Hansl code:
<hansl>
open teste.xlsx --sheet="Gretl" --rowoffset=8
</hansl>
Best regards*,*
*Henrique Andrade*
Enviado via iPhone
--
Henrique Andrade*
*
11 years, 9 months
bug in string substitution?
by Summers, Peter
Hi folks,
I'm trying to generate a matrix of "dates" that I can use on the x-axis of graphs of forecasts. I want the format to be "yyyy.q" or "yyyy.mm" for quarterly or monthly data, respectively. Things work fine for quarterly data, but something weird happens with monthly data.
I create a string with the desired format, then make the corresponding matrix entry using string substitution. So for example the string "2009.04" should just become the "number" 2009.04. However, the string substitution results in the second "decimal" being truncated, so I get 2009.0 instead of 2009.04, or 2009.1 instead of 2009.12.
The attached script illustrates the problem.
Thanks in advance for any advice!
PS
[Description: cid:logoimg]
11 years, 9 months
trivial documentation bug
by Stefano Fachin
I bumped into a trivial documentation bug in 1.9.9CVS, italian version.
I know it's old, but I'm lazy at updating programs.
The explanation of iminr and imaxr is the essentially the same, namely
"Returns the column indices of the maxima of the rows of X" and "Returns
a vector containing the column indices of the maxima of the rows of X"
whereas obviously one deals with the maxima and the other with the minimums.
... it's nice to find something wrong with Gretl occasionally, just to
avoid drifting into personality cult :-). Thanks a lot Allin and Jack.
Stefano
--
_________________________________________________________________________
Stefano Fachin
Professore Ordinario di Statistica Economica
Dip. di Scienze Statistiche
Università di Roma "La Sapienza"
P.le A. Moro 5 - 00185 Roma - Italia
Tel. +39-06-49910834
fax +39-06-49910072
URL http://w3.uniroma1.it/fachin/
11 years, 9 months
different IRFs using VAR and VECM
by Artur Tarassow
Hi all,
I am a bit confused about the following. I estimated following two
models which give exactly the same results. The first one uses the
VECM command and the second one the VAR command. For both I compute
the IRF(1,1) but they look different, which should not be the case,
or? Or do I miss something here ...
<hansl>
open denmark
list Y = LRM LRY IBO
scalar lag = 4
scalar rnk = 1
# VECM
vecm lag rnk Y --crt
matrix sigma_vecm = $sigma
matrix oirf11_vecm = irf(1,1,0)
# The same using the var command
matrix ECM = $ec
list eclist = null
loop i=1..cols(ECM) -q
series ECM$i = ECM[,i]
eclist += ECM$i
endloop
list EXO = eclist #const #|| exo2 || eclist
scalar varlag = lag-1
var varlag diff(Y) ; EXO
matrix sigma_var = $sigma
matrix oirf11_var = irf(1,1,0)
#Print-out
print sigma_vecm sigma_var
OIRF = oirf11_vecm ~ oirf11_var
gnuplot 1 2 --with-lines --time-series --matrix=OIRF --output=display
<\hansl>
Artur
11 years, 9 months
Sequential restrictions in systems
by Sven Schreiber
Hi,
it's unclear to me whether (and if so, how) it's possible to restrict
simultaneous-equations system sequentially. What I mean is something
like this:
<pseudo-hansl>
toberestricted <- system
...
end system
estimate toberestricted method=ols # unrestricted
restrict
b[1,1] = 0 # for example
end restrict
estimate toberestricted method=sur # one restriction applied
restrict
b[2,1] = 0 # again, an example
end restrict
estimate toberestricted method=sur
</pseudo-hansl>
What I would like is that the last estimation command takes into account
both previously defined restrictions, but that the test just evaluates
the last restriction, relative to the system with the previously imposed
first restriction.
(There is a similar mechanism in the VECM context when restricting the
cointegration relations; in the GUI there is a tickbox for deciding
whether to test all restrictions, or just the last ones.)
Thanks,
sven
11 years, 9 months
Interaction between two variables
by Juan S. Sánchez Oliver
Dear Gretl users,
I'm trying to introduce an interaction in the next script.
ols SPEC const Var1 Var2 COV1 COV2 COV3 COV4 COV5 COV6 COV7
I like to introduce the interaction between Var1 and Var2. What is the
correct code?
Var1 is a two-levels factor and Var2 is a continuous predictor. SPEC is the
response variable and COV1...7 are control covariates.
Thank you very much.
Regards
(And I'm sorry, my English is bad)
Juan
11 years, 10 months
issue with smpl command
by Artur Tarassow
Hi all,
in the past the following worked, but it does do so in the recent version
anymore:
<hansl>
smpl 1972:1 1978:3
string first = "1972:1"
string last = "1978:3"
scalar Tdiff = 2
smpl @first @last-Tdiff
<\hansl>
What would be a simple alternative here?
Cheers,
Artur
11 years, 10 months
installation on mac 10.7
by Paul.Anand
Can anyone help?
I have downloaded and opened friends (to some? location).
I have downloaded and created a dmg file but when I double click - nothing very much happens.
Any suggestions how I can get this going?
Thanks
Paul
--
The Open University is incorporated by Royal Charter (RC 000391), an exempt charity in England & Wales and a charity registered in Scotland (SC 038302).
11 years, 10 months
SVAR coefficients
by Gabriela Nodari
Dear gretl users,
I was wondering whether it is possible to restrict the coefficients of the
estimated structural VAR EX-POST (the estimation).
In practice, I am trying to do a counterfactual analysis and I want to
compare IRFs, by restricting one variable to not respond to an underlying
shock at ALL horizons.
Suggestions?
Thanks in advance,
Gabriela Nodari
11 years, 10 months
