If your computer will support it, try and change max workspace to 512 - we
had a date set with over 50 variables and lots of missing data and were able
to get it to work eventually. That was a crucial initial step. We also
received some crucial advice from Kenneth Scheve about, I believe as I can
remember it (I dont have my notes in front of me) about reducing /
increasing the "priors" that prevented the program from going into infinite
loop.
----- Original Message -----
From: "Gary King" <king(a)harvard.edu>
To: "Zoua M. Vang" <vang(a)fas.harvard.edu>
Cc: <amelia(a)latte.harvard.edu>
Sent: Wednesday, March 17, 2004 2:41 AM
Subject: Re: [amelia] Memory Probs
I suggest you try it with fewer variables. the number of parameters (and
hence memory requirements) increases fast with the number of variables.
Gary
: Gary King, King(a)Harvard.Edu
http://GKing.Harvard.Edu :
: Center for Basic Research Direct (617) 495-2027 :
: in the Social Sciences Assistant (617) 495-9271 :
: 34 Kirkland Street, Rm. 2 HU-MIT DC (617) 495-4734 :
: Harvard U, Cambridge, MA 02138 eFax (617) 812-8581 :
On Tue, 16 Mar 2004, Zoua M. Vang wrote:
> Hello,
>
> While running Amelia, I get this message:
>
> C:\GAUSS\SRC\VPACK.SRC(96) : error G0030 : Insufficient workspace memory
> Currently active call: VPUT [96]
>
> I tried changing the memory allotted in the "max_workspace = XX" line in
the
> GSRUN.CFG file. However, even after changing the
memory to 256, Amelia
(for
> Windows) still dies on my during the 3rd stage.
>
> How and where can I change the memory so that Amelia will run through
all 4
stages?
Thanks for your assistance.
Zoua
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Whoever you are, no matter how lonely,
the world offers itself to your imagination
--Mary Oliver
Zoua M. Vang
Ph.D. Candidate
Department of Sociology
Harvard University
509 William James Hall
33 Kirkland Street
Cambridge, MA 02138
vang(a)fas.harvard.edu