R help - Jan 2013 - Finding predicted probabilities and their confidence intervals for a logit model

I want to construct a logit model, plot the probability curve with the
confidence intervals, and then I want to
print out a data frame with the predictor, response value, predicted value,
the low ci predicted value, and the
high ci predicted value. So it should look something like:

value  low_ci   prob   hi_ci
5        0.10      0.12   0.13
6        0.11      0.13    0.16
7        0.13      0.15    0.17
....

Here's some sample data, the glm model, and a plot using the effects. I
have the plot, I
just need the presided data.

dt <- data.frame(won=c(rep(1,50),rep(0,50)), value=c(rnorm(100)))

mod <- glm(won ~ value, data=dt, family="binomial")
mod

library(effects)
plot(effect("value", mod), rescale.axis=FALSE, multiline=TRUE)


Help!

Thank You!

	[[alternative HTML version deleted]]

Hi,

I think you are looking for

as.data.frame(effect("value", mod))

Best,
Ista

On Tue, Jan 29, 2013 at 6:08 PM, Abraham Mathew <abmathewks at gmail.com>
wrote:
> I want to construct a logit model, plot the probability curve with the
> confidence intervals, and then I want to
> print out a data frame with the predictor, response value, predicted value,
> the low ci predicted value, and the
> high ci predicted value. So it should look something like:
>
> value  low_ci   prob   hi_ci
> 5        0.10      0.12   0.13
> 6        0.11      0.13    0.16
> 7        0.13      0.15    0.17
> ....
>
> Here's some sample data, the glm model, and a plot using the effects. I
> have the plot, I
> just need the presided data.
>
> dt <- data.frame(won=c(rep(1,50),rep(0,50)), value=c(rnorm(100)))
>
> mod <- glm(won ~ value, data=dt, family="binomial")
> mod
>
> library(effects)
> plot(effect("value", mod), rescale.axis=FALSE, multiline=TRUE)
>
>
> Help!
>
> Thank You!
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.