The first one, i.e., "mi" is a bug and we need to fix that. We will try to
fix it in the next release.
For the second error, unfortunately, the two stage least squares (as
implemented by systemfit) do not take weights.
Kosuke
On Tue, 4 Mar 2008, Casey Klofstad wrote:
I'm trying to run twosls. Neither the mi nor the
weights command works.
1. "mi"
Here is the code I ran:
fml <- list(mu1 = civtotw3 ~ civtotw1, mu2 = civtotw2 ~ civtotw1 +
treat, inst = ~ civtotw1 + treat)
z.out <- zelig(formula = fml, data = mi(ds1, ds2, ds3, ds4, ds5),
model = "twosls")
It all runs without error, but when I run summary(z.out), I get this:
"Error in apply(coef1, 1, mean) : dim(X) must have a positive length"
But, if I run the command on each imputed data set individually, the
code above works.
2. "weights"
Here is the code I ran:
"Error in systemfit.control(...): unused argument(s) (weights =
"weights")"
If I take out the "weights" command, the code works fine.
Thanks for the help!
-c
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options:
http://lists.gking.harvard.edu/?info=zelig
Zelig program information:
http://gking.harvard.edu/zelig/