CVS news
by Allin Cottrell
A few changes in response to recent comments:
* Andreas's request for ceiling and rank-order functions: There's
now a "ceil" function in genr; a rank-order function will follow.
* Sven: spurious error message on supplying a null list as the
target of a "foreach" loop. Fixed.
* Sven: variables disappearing after calling a user-defined
function that takes a list as an argument. I have fixed one
problem in this area. I think that what I fixed was the main --
and perhaps the only -- error, but I need to do some more analysis
and testing to make sure.
There's a new Windows snapshot in place.
Allin.
13 years, 5 months
Ang. Re: [Gretl-users] New rank function for genr
by andreas.karlsson＠ltv.se
cottrell(a)wfu.edu @ INTERNET skrev 2007-05-23 13:45:03 :
> On Tue, 22 May 2007, andreas.karlsson(a)ltv.se quoted:
>
> > > As for "mode()", I'm all for it, provided we find a sensible > >
> way to handle multi-modal variables.
>
> What behaviour do you think is sensible when a variable does not
> have a unique mode?
>
> x = {3,3,3,4,4,4,5,5,5}
Well, it is of course a problem with multi-modal variables. This would e.g.
result in mode = 3, 4, and 5. On second thoughts, maybe it is not that
important to include the calculation of the mode in gretl. At least not
until a sensible way to handle multi-modal variables is found.
Med vänliga hälsningar / Best regards
Andreas Karlsson
13 years, 5 months
Cannot open session file
by Max Bölhoff
When I first started using gretl last autumn I had an issue that one day I could no more open the session file of a project I had been working on over the course of several days. Now, with version 1.6.5 (Windows), exactly the same has happened again, and when I look through the files in the Documents and Settings folder as well as the one where I saved the session file, only the latter has actually been saved when I last did so, but the supporting files in the Documents and Settings folder are old and imcomplete, meaning that several save commands have failed?! The error message I am receiving when trying to open the session file is "Couldn't open session.xml" which is one of the files in the Documents and Settings folder, and which, as just said, has not been updated throughout several save requests. So tell me, what have I done wrong that this could happen? As much as I appreciate this free software, and even prefer to some of the lesser commercial packages for my
particular purposes, I cannot afford to lose several days of work being in a commercial environment. Thanks for any assistance provided.
___________________________________________________________
Yahoo! Mail is the world's favourite email. Don't settle for less, sign up for
your free account today http://uk.rd.yahoo.com/evt=44106/*http://uk.docs.yahoo.com/mail/winter07....
13 years, 5 months
Re: Ang. [Gretl-users] nonlinear IV (fwd)
by Riccardo (Jack) Lucchetti
On Tue, 22 May 2007, Sven Schreiber wrote:
> Riccardo (Jack) Lucchetti schrieb:
>
>>
>>> svetosch(a)gmx.net @ INTERNET skrev 2007-05-22 13:12:59 :
>
>>>> I'm interested in applying nonlinear instrumental variables, like in
>>>> section 7.6 of the Davidson/MacKinnon book.
>>>>
>>
>> I don't have the book with me atm. I suppose it can be done, but would I
>> be very mistaken in thinking that the gmm command already does what you
>> want?
>
> No and yes; no you're not mistaken that one could use the gmm command,
> but yes you're mistaken that that's what I want :-) I was hoping to
> exploit the properties of the special case of nonlinear IV, and spare me
> some thinking about the correct formulation of orthogonality conditions.
> For example, in Eviews I believe it's possible to specify a nonlinear
> equation and a list of instruments and you're done. (I know that Eviews
> has a bad reputation with numerical optimization, that's one of the
> reasons why I'm looking for alternatives.)
Could you be more specific on this? I mean, suppose you want to estimate
something like y_t = f(x_t, \theta) + \epsilon_t via NLIV. What you can do with
GMM is write the orthogonality condition as y_t - f(x_t, \theta) and go from
there.
A little example script follows, where I set up a simple example of NLIV
estimation in a measurement error textbook situation:
#---- begin example script ----------------------------------
nulldata 100
set seed 23897
genr e = normal()*0.1
genr xstar = 3 + uniform()
genr x = xstar + e
genr y = sqrt(xstar) + normal() + e
genr z1 = xstar + normal()
genr z2 = xstar + normal()
scalar theta = 0.25
matrix W = I(2)
list Z = z1 z2
gmm
series e = y - x^theta
orthog e ; Z
weights W
params theta
end gmm --iterate
#---- end example script ------------------------------------
Is the problem you have at hand too hard to tackle via the above approach?
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti
13 years, 5 months
New rank function for genr
by andreas.karlsson＠ltv.se
r.lucchetti(a)univpm.it @ INTERNET skrev 2007-05-22 16:10:30 :
[snip]
> ceiling() whould be easy to do. However, with a bit of dexterity you can
> coerce the "int" function to do what you want; for example:
>
> nulldata 100
>
> x = int(uniform()*6)/2
> f = int(x)
> c = (x-f) ? f+1 : f
It surely would be easier to use genr y = ceiling(x)
> As for "mode()", I'm all for it, provided we find a sensible way to
handle
> multi-modal variables.
Also, when we are discussing new functions, a "rank()" function for the
"genr" command, that gives the ranks of the values in a vector, with ties
resulting in ranks being averaged, would be useful.
Regards,
Andreas
13 years, 5 months
Ang. Re: [Gretl-users] New rank function for genr
by andreas.karlsson＠ltv.se
r.lucchetti(a)univpm.it @ INTERNET skrev 2007-05-22 18:25:21 :
> On Tue, 22 May 2007, Sven Schreiber wrote:
> > (Aside from that, I wouldn't call it rank because of the rank concept
in
> > linear algebra.)
>
> Sven is right. We already have the rank() function as a matrix function.
> Changing its behaviour if the argument is a series is feasible, but I'm
> wondering if the usefulness of rank() for series outweighs or not the
> potential confusion that may arise. We probably should leave the decision
> to the commander-in-chief (no, not THAT one).
Well, of course the function need not be called "rank()". Call it
"ranking()" or "rankorder()" or whatever. The name is not important but
what it performs, namely assigning ranks to the values in a series.
Regards,
Andreas
13 years, 5 months
Re: [Gretl-users] adf test retreival question
by a b
Franck Nadaud a écrit :
Dear listers, greetings from Paris !
I would like to submit a question relative to a script ...
I'm not yet very familiar with matrix but instead of using a matrix, try to create two series -- a_vect and p_vect (such as <genr series a_vect>-- and you can join both of them in a matrix after exiting the loop ! That'ss what I do at least at it works!
Bien à toi!
Artur
---------------------------------
Stockage illimité de vos mails avec Yahoo! Mail. Changez aujourd'hui de mail !
13 years, 5 months
adf test retreival question
by Franck Nadaud
Dear listers, greetings from Paris !
I would like to submit a question relative to a script where I need to create
a matrix holding the $test and $pvalue of succesive adf tests on various
series pertaining to the set of 22 french regions.
In my script i loop on the regions, then restrict to each region in succession
and compute the adf tests of my series.
The tests results matrix is updated successively, after each test. However,
gretl does not excute the commands because it says :
"no formula in given in genr" (translated from french)
I do not understand why it does not work
the code looks like :
adf 1 x1
genr a=$test
genr p=$pvalue
matrix A($i,1)=a
matrix A($i,2)=p
adf 1 x2
etc
it seems their is a problem with genr and passing the values to the matrix
entries.
any advice would be very very welcome !
cheers
Franck
--
Franck Nadaud
Economiste
CIRED
UMR 8568 CNRS - EHESS
45 bis avenue de la Belle Gabrielle
94736 Nogent-sur-Marne Cedex
TEL 33-1-43-94-73-94
FAX: 33-1-43-94-73-70
MOB: 06-07-39-92-75
France
13 years, 5 months
Ang. [Gretl-users] nonlinear IV
by andreas.karlsson＠ltv.se
svetosch(a)gmx.net @ INTERNET skrev 2007-05-22 13:12:59 :
> Hi fellow gretl users,
>
> I'm interested in applying nonlinear instrumental variables, like in
> section 7.6 of the Davidson/MacKinnon book.
>
> AFAIK it's not directly possible in gretl, but what about using the
> generic maximization routine that was recently added to gretl (IIRC)? Is
> that ready yet?
A generic minimization (or maximization) command in gretl would really be
welcome. What algorithm will it use? Newton-Raphson?
BTW, when talking about gretl commands, a function "mode()" that calcualtes
the mode, i.e. the value that has the largest number of observations, and a
function "ceiling()" that computes the ceiling function, would be a welcome
addition to be used with the "genr" command.
Med vänliga hälsningar / Best regards
Andreas Karlsson
13 years, 5 months
nonlinear IV
by Sven Schreiber
Hi fellow gretl users,
I'm interested in applying nonlinear instrumental variables, like in
section 7.6 of the Davidson/MacKinnon book.
AFAIK it's not directly possible in gretl, but what about using the
generic maximization routine that was recently added to gretl (IIRC)? Is
that ready yet?
Or would it be easy to extend gretl's NLS code to do it natively? (For
people familiar with gretl's interiors, I mean.)
Thanks,
Sven
13 years, 5 months
