[Gretl-users] problem with system estimation

[Allin Cottrell]

On Tue, 11 Mar 2008, Franck Nadaud wrote:

> I just installed the winsnapshot to keep up to date but... I 
> discover some rather disapointing problem with system 
> estimation.
> 
> I tried to run my scripts for AIDS and QUAIDS estimation that 
> ran perfectly. However, under current snapshot none of my system 
> commands run, GRETL just says : data error...

Thanks for pointing this out.  It was due to a bug in the new 
mechanism for generating forecasts with standard errors from 
equation systems.  I think it should now be corrected in CVS and 
the current Windows snapshot.  (Also, I've now added some of your 
scripts to my gretl test rig, so hopefully this sort of breakage 
will not occur again.)

> I also noticed something else : at start GRETL takes a lot 
> longer to load into memory and appear ready.

I can't think of anything that has changed lately in gretl that 
would account for that.  

-- 
Allin Cottrell
Department of Economics
Wake Forest University, NC