Heather and others,
I have had the same problem imposing a range prior on a variable. I
tried to place a range prior of 0 to 2 with varying confidence
coefficients. I received the same error message that disappeared when
the prior was removed. Also, after setting the range, the mean in the
observational priors window was 2, the maximum of the range.
I did not find a solution, but wanted to let you all know that I
experienced the same problem.
Erin Saunders
Portland State Univ.
On 3/9/07, Heather Stoll <hstoll(a)polsci.ucsb.edu> wrote:
Dear all,
I earlier posted a message about what struck me as something odd about
how AmeliaView is treating range priors. Subsequently I've tried
imputing the same data set (using the same parameters) in R itself.
(I'm running the most recent versions of both R and Amelia on a
Windows XP system.) I've tried many "tweaks" to the entire-variable
range priors (exactly which variables receive a range prior; the
minimum and maximum values; the confidence coefficient), and no matter
what I do, whenever I include range priors, Amelia returns an error of
the following sort:
Error in chol(copy.theta[c(FALSE, m[ss, ]), c(FALSE, m[ss, ])]) :
the leading minor of order 3 is not positive definite
Amelia runs without a hitch for this data set once I eliminate the
range priors.
Any help would be much appreciated.
Best,
Heather Stoll
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=amelia
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=amelia