Re: [Gretl-devel] Re: [Gretl-users] matrix declaration error
Allin Cottrell schrieb:
>
> But what are those things you guys added?
We added a couple of functions lately, namely
nullspace(X)
which gives the right null space for matrix X, and
svd(X, &U, &V)
which returns the singular values of X as a vector, and provides the
left singular vectors of X (in U, if non-null) and/or the (transpose of)
the right singular values (in V, if non-null).
We've also fixed up ".^" (element-wise exponentiation) so that it works
in the same flexible way as ".*", and have added operators ".+",
".-"
and ".=" that work in the same way.
That's all very nice!
> Apart from that I'm a bit hesitant because porting to gretl script
> means abandoning object-orientation. So the structure would be quite
> different, and therefore maintenance effort of the parallel python and
> gretl programs would be a bit annoying.
Sorry, I haven't had time to take a close look at your py4gretl code
yet; I might have some suggestion to make after doing that.
Actually I believe that I now understand the capabilities and
limitations of gretl script pretty well, and to be honest I think it
would be too much effort to port _everything_ to gretl script.
However, with respect to some of the more straightforward stuff (like
just doing the standard Johansen procedure, but allowing for restricted
exogenous variables), I am more optimistic!
cheers,
sven