Re: [Gretl-users] Hide or show lagged variables?
On Sat, 14 Jan 2006, Allin Cottrell wrote:
Makes sense. I too have a vague recollection of how PcGive does it,
but -- apart from the issue of gretl's regressor selection
sub-window becoming overcrowded -- I feel that gretl's selector is
easier to use.
Anyway, if anyone has any brilliant ideas regarding the ideal
interface, please post them!
I have an idea, which may not be brilliant, but at least is new.
What if the user had the possibility of selecting lists together with
regressors[1]?
If someone stubbornly refused to use the console or scripts, we could have
a menu item like "Create list with lags" (possibly right-clickable). Then
the list containing the wanted lags appears as if it was a single series,
and it can be included in the regression.
Note that in this case it wouldn't be necessary even to create the lagged
vars themselves, since inclusion of such list in an estimation command
simply acts as a "placeholder", so gretl knows we want lagged values too.
As a side benefit, this would also solve the problem I mentioned in my
previous message on modifying a vatriable after the lags are generated.
Now, 2 questions:
a Does anyone like the idea?
b if(a==TRUE) {
Would it be hard, Allin?
}
[1] Brief note for the benefit of non-regulars: gretl has had lists for
some time now; maybe not everyone is aware of this, because there's no
list-related element in the GUI, but you can use lists in script like this
list zlags = z(-1 to -4)
or even
list zlags = lags(z,4)
and then do
ols y const zlags
which would be equivalent to
lags 4 z
ols y const z_1 z_2 z_3 z_4
there's more to lists, but you can read the manual for that ;-)
Riccardo `Jack' Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
jack(a)dea.unian.it
http://www.econ.univpm.it/lucchetti