Re: wishes list bis
by andreas.karlsson＠ltv.se
my_gretl(a)yahoo.fr @ INTERNET wrote 2007-05-28 09:06:08 :
> Would it be possible please, to save a matrix as a .gdt file
> for further use?
> When I saved a matrix as .txt file and tried to import it into
> Gretl, the result was a really mismatched and unusable file.
>
> Best,
> Artur
>
I do agree that it would be useful to have the possibility to save a matrix
as a .gdt file for further use.
In addition to this, it would also be great to have the possibility to
export a matrix in CSV, GNU R, Octave, JMulTi or PcGive file format.
Regards,
Andreas
16 years, 10 months
Heckit
by Gordon Hughes
I believe that the Stata option - i.e. consistency with the ML
estimator - should be the default option because (a) it is then clear
what is going on, and (b) it is possible to compare the results for
the two step estimator with the ML estimator. (I have a general
preference for the ML estimator because the two step estimator often
seems to generate spuriously low standard errors on the
coefficients.) Further, the user should be discouraged from using
the two step estimator with non-matching missing data for the reason
that you give.
But, there is an important secondary consideration. Certain types of
survey design can generate systematic patterns of missing data - this
would include the answers to questions that are only asked if someone
is in the labour force or bought certain goods in the last month. In
such cases, exclusion of all observations with missing data can
seriously compromise the possibility of estimating a model reliably
if the pattern of questions asked/answers is correlated with the
selection probability.
There is an alternative way of analysing such data. It is
straightforward to let a user construct their own two step estimator
with different missing data as follows: (a) estimate the probit model
for selection and generate the Mills ratio as a new variable; then
separately (b) estimate the OLS equation including the Mills ratio as
a dependent variable. The corrections are not so difficult and
anyone following this route explicitly should know what they are
doing and can be warned in the documentation. All that is needed is
an option in the probit model to generate the Mills ratio as a
post-estimation variable. I haven't checked whether it is there
already but it could easily be added.
Gordon Hughes
16 years, 10 months
Ang. Re: [Gretl-users] wishes list
by andreas.karlsson＠ltv.se
my_gretl(a)yahoo.fr @ INTERNET skrev 2007-05-28 11:46:18 :
> Bulent Miran a écrit : Would it be possible to
> get p values in logit/probit outputs?
[snip]
> I'm afraid not. As long as I'm aware, Gretl has not an internal
> variable grasping the p-value of estimators. But you can always
> compute it yourself.
> For example, after executing "logit" or "probit" command you can
> add the following code :
>
> genr p_val = pvalue(t, $df, $coeff(X)/$stderr(X))
>
> and you'll get what you need
>
> Artur
I the p-values can be included as default in logit/probit outputs.
Regards,
Andreas
16 years, 10 months
Re: [Gretl-users] wishes list
by a b
Bulent Miran a écrit : Would it be possible to get p values in logit/probit outputs?
B Miran
----- Original Message -----
From: a b
To: gretl-users(a)lists.wfu.edu
Sent: Monday, May 28, 2007 10:02 AM
Subject: [Gretl-users] wishes list bis
Would it be possible please, to save a matrix as a .gdt file for further use?
When I saved a matrix as .txt file and tried to import it into Gretl, the result was a really mismatched and unusable file.
Best,
Artur
---------------------------------
I'm afraid not. As long as I'm aware, Gretl has not an internal variable grasping the p-value of estimators. But you can always compute it yourself.
For example, after executing "logit" or "probit" command you can add the following code :
genr p_val = pvalue(t, $df, $coeff(X)/$stderr(X))
and you'll get what you need
Artur
---------------------------------
Stockage illimité de vos mails avec Yahoo! Mail. Changez aujourd'hui de mail !
16 years, 10 months
wishes list bis
by a b
Would it be possible please, to save a matrix as a .gdt file for further use?
When I saved a matrix as .txt file and tried to import it into Gretl, the result was a really mismatched and unusable file.
Best,
Artur
---------------------------------
Stockage illimité de vos mails avec Yahoo! Mail. Changez aujourd'hui de mail !
16 years, 10 months
Re: [Gretl-users] Heckit
by a b
Riccardo (Jack) Lucchetti a écrit : I'm working on the Heckit estimator, and I need some feedback on the treatment of missing observations. For the two-step estimator, Stata uses matching samples. What should WE do?
From my point of view, let us do what Stata does!
Have you checked other econometrics softwares? I'm not very familiar with Heckit estimator but I guess RATS provides a procedure for it.
Best, Artur
---------------------------------
Stockage illimité de vos mails avec Yahoo! Mail. Changez aujourd'hui de mail !
16 years, 10 months
Re: [Gretl-users] Linux versions of gretl
by Gordon Hughes
My point was not really specific to the difference between versions
1.6.2 and 1.6.5, since I have used gretl on Windows for some time and
can continue to do so. Further, I don't doubt that it is possible to
rely upon some versions of Linux which are directly compatible with
the versions relied upon by gretl's developers or packagers. But,
Linux is a broad church with an increasing of users who look for a
standard desktop distribution - most of which are conservative in
updating core libraries.
However, I have found that trying to update various libraries to
resolve gretl dependencies has caused me serious problems in several
versions of Linux including Debian, Suse & RH derivatives. What I
have learned is that you go beyond the versions that are provided by
the relevant repositories at your peril and in many cases
distributions rely upon releases of key libraries that are several
generations out of date. Secondly, there can be important but
unexpected problems even when the dependencies appear to have been
resolved. For example, I have just discovered that I can't run
version 1.6.2 properly because my installed version of LAPACK has
unresolved symbols to the GNU Fortran library. This is irritating
but not disastrous for me because I used to design and write
econometric software many years ago, so that I can track down the
missing libraries though I would prefer not to have to spend the time
on it. Also I have a Windows virtual machine on my Linux system on
which I can easily run gretl, though that rather defeats the point of
the exercise.
The key issue is that this kind of difficulty is likely to put off
95% of potential or actual users of gretl on Linux. Administrators
of centrally managed computing facilities hate to deal with such
matters. That is the reason for considering a statically linked
version for Linux - to widen the potential group of users beyond the
relatively small group of highly motivated cognoscenti.
Gordon Hughes
16 years, 10 months
Heckit
by Riccardo (Jack) Lucchetti
Hello all,
I'm working on the Heckit estimator, and I need some feedback on the
treatment of missing observations.
Some background first: suppose we want to estimate a model
y = X \beta + u (1)
but we only observe y if another variable d equals one. Assume that
P(d=1) = \Phi(Z \gamma) (2)
As you all know, there are two way to estimate \beta:
a) Estimate a probit model first for d, compute the Mills ratios and stick
them into (1), which you estimate by OLS. With an appropriate correction
of the OLS std. errors, what you get is the "two-step" estimator.
b) Estimate \beta, \gamma and the correlation parameter together by
maximum likelihood. This is arguably preferable.
Now suppose you have some missing observations in X, in Z or both (far
from unusual in large micro datasets). Obviously, for the ML estimator you
can only use the observations that have no missing values for any of the
variables.
With the two-step estimtors, however, you may have different samples for
the two equations (1) and (2): if there are missing data in X only,
nothing forbids you from estimating (2) on the full sample and then (1) on
the subset for which you actually have data.
Would this be good or bad? The answer I gave myself so far is that on one
hand, if you use all the data for (2), you end up with better estimates of
\gamma, which in turn gives you better estimates of the Mills factor and
hence of \beta. This, of course, assuming that the probabilistic mechanism
which dictates which rows of X are missing is independent of everything
else; otherwise, this could be a VERY bad idea.
For the two-step estimator, Stata uses matching samples. What should WE
do?
Comments welcome. And, oh, before you say "let the user choose", let me
just say that yes, this is a possibility, but then, what should the
default behaviour be?
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti
16 years, 10 months
Re: [Gretl-users] wishes list
by a b
Ignacio Diaz-Emparanza a écrit :
1- Add a new command "rates" and a new item to the "Add" menu (for time series), which computes Y/Y(-1)-1.
It's a very good idea.
Some time ago I had to handle data from World Development Indicators (WB) and computed the growth rates over a period by script (and loops). It will be great if Gretl could do the same by a build-in command, using WB's method (log regression on time).
By the way, how can I add a trend to a white noise ? I guess by "genr y = normal() + time" ???? And what about a seasonal trend ?
Best,
Artur
---------------------------------
Stockage illimité de vos mails avec Yahoo! Mail. Changez aujourd'hui de mail !
16 years, 10 months
arch forecasting
by Chris
Hi,
I'm puzzling over this arch stuff trying to figure out how to use the
results to forecast. When I estimate the arch model I get a bunch of alphas
estimated and the usual output for coefficients. To forecast do you just use
the coefficients on the X's and multiply them by your new values to get a
forecasted Y? Do I have to do anything with the alphas? Reading my copy of
Greene, they just seem to be forcing structure on the error term and
estimating that structure so I'm just not certain if I have to use them
somehow to get a forecasted value or not.
Thanks,
Chris
16 years, 10 months
