Re: [Gretl-devel] bug when using gmm
Riccardo (Jack) Lucchetti schrieb:
Fortunately, it isn't.
When you do 2 or more steps in GMM, the weights matrix gets updated.
Hence, in your script the second GMM blocks starts with the weights
matrix out of the first one. If you insert the statement
matrix dnew2weights = inv(dnew2mat'dnew2mat)
between the two gmm blocks, you get that the second estimate is
precisely equal to what would have been if you had skipped the first gmm
block.
First of all, thanks a lot for the quick explanation!
However, I'm wondering whether the updated weight matrix shouldn't
rather be stored in some internal gretl variable instead of what the
user passed as the *initial* weights. Are there any other cases where a
user-defined variable is implicitly altered by gretl?
IMHO at least it should be stated explicitly in the manual -- but I
confess I haven't checked and I can very well imagine that it is already
mentioned there.
thanks,
sven