I'm running a rare events logit model in R using the Zelig package and am
getting some of the oddest error messages that I can't figure out.
y = rnorm(100)
x = c(rep("0",1), rep("1",99))
d = data.frame(won=x, bid=y)
d
mod1 <- zelig(y~x, model="relogit", data=d, tau=1/100,
case.correct="prior", bias.correct=TRUE, robust=TRUE)
> mod1 <- zelig(y~x, model="relogit", data=ndat, tau=1/100,
case.correct="prior", bias.correct=TRUE, robust=TRUE)
<simpleError in family$linkfun(mustart): Value 1.16885 out of range (0,
1)>
Error in eval(expr, envir, enclos) : object 'form' not found
In addition: Warning messages:
1: In model.matrix.default(mt, mf, contrasts) :
variable 'x' converted to a factor
2: In eval(expr, envir, enclos) : non-integer counts in a binomial glm!
3: In value[[3L]](cond) :
There was an error fitting this statistical model.
Object form isn't found, but there is no mention of it in any of my code,
so I don't know what the issue is.
> str(d)
'data.frame': 100 obs. of 2 variables:
$ won: chr "0" "1" "1" "1" ...
$ bid: num 1.46 1.84 -0.13 -1.16 1.01 ...
Playing with the modes of each variable has proved to be fruitless also.
I've looked through the Zelig help list and lot of people seem to have this
issue, but no one had posted an answer or resolution.
--
*Abraham Mathew
Statistical Analyst
**720-648-0108*
*abmathewks@gmail.com
*
*Twitter <https://twitter.com/abmathewks>
**LinkedIn<http://www.linkedin.com/pub/abraham-mathew/29/21b/212/>
**Blog <https://mathewanalytics.wordpress.com/>
**Tumblr<http://fitnessplusfashion.tumblr.com/>
*
[[alternative HTML version deleted]]
