These questions need to be addressed to either Amelia or cem list.
Thanks,
Kosuke
--
Department of Politics
Princeton University
Hi,
I'm trying to use cem after Amelia. I'm getting sensible results, but also
various error/warning messages. Any assistance/clarification about them would be
appreciated. Below I'll describe the two scenarios in which I'm getting three
different error/warning messages.
I realise this is a Zelig list, but I'm hoping that questions about Amelia and/or
cem are answerable here--if not, please let me know.
Many thanks,
Malcolm
===============
FIRST, without using Amelia, in some instances when I call "att" (within cem
package), I'm getting:
mod <- att(mat, depvar ~ treated +
as.factor(control1) + as.factor(control2), model="logit", data = dat)
Warning messages:
1: In eval(expr, envir, enclos) :
non-integer #successes in a binomial glm!
2: In predict.lm(object, newdata, se.fit, scale = 1, type = ifelse(type == :
prediction from a rank-deficient fit may be misleading
Comment: My outcome variable is definitely only zeros and ones, so the complaint that it
contains something else is weird. From searching around on the web, my impression is that
including weights in a binomial glm can generate this warning message. And my
understanding is that cem uses weights when calling "att". So is this an
apparently alarming but actually harmless glitch in att? I don't have any ideas about
the second warning.
===============
SECOND, in other instances, after multiply imputing missing data with Amelia, and
combining the five imputations into an object "impute", I'm getting:
mod2 <- att(mat2, depvar2 ~ treated, data =
imputed, model="logit")
summary(mod2)
Treatment effect estimation for data:
NULL
Logistic model estimated on matched data only
Coefficients:
Error in symnum(pv, corr = FALSE, na = FALSE, cutpoints = c(0, 0.001, :
'x' must be between 0 and 1
Comment: I can get details about "mod2" by calling "str(mod2)", but
that's obviously cumbersome and I'm curious why I can't use the usual
"summary" function.
-
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/
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options: