the number of parameters in your model is 35*38/2=665, which is not a lot
less than your n. (If you have set _AMnoms at all, then the number will
even be higher). I'd start by doing an Amelia run with say 5 variables,
just to get a feel for it. Then add more. But I think you should
probably stop before 38.
Gary King
: 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 (928) 832-7022 :
On Mon, 12 Aug 2002 abegunded(a)who.int wrote:
Dear Dr. Gary King,
I use Amelia for windows (version: 2.01, 3/19/2001), quite well with
analysing data. However, on this occasion it fails to run as it keeps
tripping off the screen before completing the imputation. Herewith the log:
*********************
Amelia Begins
*********************
Inputs valid, beginning EMis...
******************************************
* Stage 1 of 4: EM (Finding Mode)...
******************************************
Using conditional model
Iteration Number:
C:\GAUSS\SRC\SWEEP.SRC(102) : error G0048 : Matrix singular
Currently active call: SWEEP [102]
My data consist of 995 observations with 35 variables. intervariable
missingness is estimated to below 10%.
_AMempri was set to between 5 and 10 at various tries. _AMfully, _AMords
were appropriately set and _AMsn were set to the value set for _AMmpri.
I have reinstalled recent copies yet the problem persists.
I would appreciate your suggestions to overcome this problem.
Regards,
Dr. Dele Abegunde
NMH/CVD
World Health Organisation
20 Avenue Appia
CH - 1211 Geneva 27
Switzerland.
Tel: +41 22 791 2826
Fax: +41 22 791 4151
-
amelia mailing list served by Harvard-MIT Data Center
List Address: amelia(a)latte.harvard.edu
Subscribe/Unsubscribe:
http://lists.hmdc.harvard.edu/?info=amelia