I ran 120 regressions with a script and now I want to re-run them but only
with the variables that were found to be *statistically important and
positive*. Do you know of any scripting methods to do this automatically so
that I won't have to edit 120 regressions by hand? Thank you for your time!
I was just wondering what estimation method is used exactly in the
Johansen_stage_1 function (the first step, VAR estimation, of the
Johansen test) of lib/var.c. From looking at the code, it seems like
it might be the 2-step Least Squares method -- but I'm not clear on
where in the code the fitted values are plugged in, it's almost as if
X is never changed between the first and second OLS. Also the comment
refers to log-likelihood.
So is it 2SLS, or is it some maximum likelihood method like FIML? (or
to estimate PSTR models:
a concentrated SSR must be minimized and the authors suggest to choose
the starting values by means of simulated annealing.
Reading the manual, I discovered that gretl is already able to use
simulated annealing to adjust the initialization in case of ML estimates
in the restrict command via the option --jitter (btw: such option is not
reported in the description of the command in the command reference).
Is there a way to use such method in gretl for initializing a generic
function to be minimized/maximized?
In comparing the following scripts, the matrix operations are much more
time-consuming tasks. Is such a behaviour normal ? I was under the
understanding that handling matrices is quicker than operations on
series x = normal()
series example = 0
loop for i = 1..$nobs --quiet
example[i] = x[i]
matrix example2 = zeros($nobs,1)
loop for i = 1..$nobs --quiet
example2[i,1] = x[i]
development economist, consultant
phone: +216 22 99 73 44
I will be out of the office starting 10/05/2011 and will not return until
Hi All, I will be out of the office on business from October 5 to Oct 7
returning to the office Oct 10, 2011. I will be cheking emails in the
evening. But, if you have an immediate concern with Cost Model, email
CM-Help(a)kp.org or call Bradley Njus at 510 625 4595.
work cell 510 435 6528
since there is new version of x-12-arima (0.3 build 192) I've build DEB
and RPM packages for AMD64 architecture (with PSP = 24 patch). Files can
be downloaded from: http://wrzosy.dyndns.info/~marcin/x12arima/
Allin, would you build i386 packages?
And the last issue: in current version RPM package has version
x12a-0.3-2 while DEP package x12a_0.3-3 which was probably the efect of
lack of "-k" switch for alien. I think we could fix it.
We're planning to release gretl 1.9.6 before long -- probably a
couple of weeks from now -- and in the meantime we'd much appreciate
it if you could help us hunt down bugs, so that we can make the best
If you have old gretl scripts to hand, please try running them and
check that they still work. And if you have a record of results,
please check that the current results agree with the old ones.
Also -- since checking the gretl GUI for breakage is not so easy as
checking scripts -- please see if you can find time to do things
using the graphical interface and look out for anything wrong.
Send your reports of bugs to this list, or use the bug-tracker at
Sourceforge if you prefer:
For now, however, please hold back on feature requests -- we're
focused on finding bugs.
Please bear in mind a couple of important points:
1) Your reports will be much more valuable if you're using gretl
that you have build yourself from current CVS, or else the current
snapshot for MS Windows
or Mac OS X.
Many bugs in the last release (gretl 1.9.5) have already been fixed.
See http://gretl.sourceforge.net/ChangeLog.html .
2) Some older gretl scripts will not work because the scripts
thenselves contain bugs. Current gretl catches (for example) the
syntax error of writing
scalar x = 2*(3 + 4)/y)
which has an extraneous trailing ')'. This was fixed just today
(2011-09-04). I was surprised at how many scripts in my collection
exemplified this error.