On Thu, 28 Jul 2011, adkins.lee(a)gmail.com wrote:
This is excellent to know. I've been using functions more to do
routine
things (e.g., model selection and computation of marginal effects) and
they are unhappy when fed the wrong type. I'm slowly coming to grips
with it and like it a lot. I think the move to more explicit types will
facilitate the expansion of functions (and bundles?).
I tend to avoid "genr". The two cases in which you definitely want to use
the specialised syntax which come to my mind right now are
a) generating a constant series, eg "series foo = 2.71"
b) making sure that a 1x1 matrix is interpreted as such, rather than a
scalar or vice versa. We've made lots of progress in handling possible
ambiguities automatically, but there are still a few grey areas and I
doubt that the issue can be resolved conlusively.
So, it don't cost you nothing to use "scalar", "matrix" and so on
instead
of "genr", but it does have its advantages (in terms of code readability
too). This, of course, IMHO.
I am very excited about the new version of "the book". I
think it will
be a significant upgrade. At least I hope so
Can't wait to see it!
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti