On Thu, 22 Apr 2010, Franck Nadaud wrote: > I finally checked my problem and here are my conclusions [snip] > 3) A first line of conclusion is thus that : first, the loops > are not under cause ; second, it seems that the system commands > do not handle colinearity in the variables as efficiently as > does ols. This seems rather serious because instead of getting > an error message gretl crashes.
Thanks for investigating! That is now fixed in CVS: if exact collinearity is encountered in estimating a system you now get an error message instead of a crash. Maybe later we can think if it's worth trying to support automatic elimination of regressors in this case, as with single-equation OLS. Allin