On Thu, 6 Aug 2009, Sven Schreiber wrote:
Allin Cottrell schrieb:
>> Hi, I have just discovered gretl. It is quite convenient that
>> insignificant regressors can be sequentially dropped from a
>> regression. However, sometimes want does not want to drop certain
>> regressors right away, even if they are insignificant (e.g. variables
>> in levels in an unrestricted error correction model). Is it possible
>> to exclude some variables from the routine? How?
> You can do this quite easily via a script. I'm not sure it's
> worth implementing as a built-in command.
sorry Allin, I don't see how, at least not in combination with the
built-in auto omit feature. (It's clear that a handmade script can be
written for a specific situation, but then it may be no quicker than
doing the sequential drop by hand.)
Agreed, a script would have to be written from scratch and could
not use the auto-omit feature.
Thinking about this some more I realized that one variant of the
request was trivial to implement. Namely, if you give a list to
the omit command along with the --auto option, the automatic
elimination is confined to the listed variables. I've added that
In that context I realized that there was a hole in "genr" with
regard to lists: you couldn't create a new list by subtraction of
one list from another (although you could subtract an individual
variable from a list).
That's now fixed, so if you're thinking in terms of a protected
subset of variables that should _not_ be omitted you can do this
sort of thing:
list X = x*
ols y 0 X
list protect = x3 x4
list omitv = X - protect
omit omitv --auto
This might be a little awkward, but then I think it could be
confusing to switch the sense of the list supplied to omit in the
case where the --auto flag is given. And we have a precedent in
several commands for "list given means operate on the listed
variables; no list given means operate on all relevant variables",
which is how omit --auto now works in CVS.