[amelia] RE: Amelia does not converge

Prof. King, Thanks for your prompt reply. I forgot to mention that I'm working with event count data rather than surveys. The two variables that most interest me, however, are the ones with high missingness... Jose Aleman PhD Candidate Politics Department 130 Corwin Hall Princeton, NJ 08544 609.937.0190 609.258.2147 -----Original Message----- From: Gary King [mailto:king@harvard.edu] Sent: Monday, May 10, 2004 5:17 PM To: Jose A. Aleman Cc: Amelia Listserv Subject: Re: Amelia does not converge the theorem only says that if you let it run it will eventually finish, but there are problems sometimes. almost all the time the problem is due to difficult or impossible datasets, such as if your panel design is entirely missing for all the variables for respondents who didn't answer one of the waves of the survey. I suggest you start with just a couple of variables that do not have much much missingness, get that to converge, and then add other variables one or a few at a time and see what happens. The problem may be entirely in one variable or set of variables, which you could then remove. Best of luck, 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 (617) 812-8581 : On Mon, 10 May 2004, Jose A. Aleman wrote: of runs completely