R help - Aug 2012 - Using the effects package to plot logit probabilities

I'm trying to run a logit model and plot the probability curve for a number
of the important predictors. I'm trying to do this
with the Effects package.


df=data.frame(income=c(5,5,3,3,6,5),
              won=c(0,0,1,1,1,0),
              age=c(18,18,23,50,19,39),
              home=c(0,0,1,0,0,1))
str(df)

md1 = glm(factor(won) ~ income + age + home,
          data=df, family=binomial(link="logit"))

summary(md1)

plot(effect("income", md1), grid=TRUE)


But I want to know how to plot a graph so that it shows the probability of
won (response) based on income (or any of the other predictors).

However, what I want to do is generate the same plot, with won don y axis
and income on x axis, but the curves showing the probabilities for age and
home.


Not seeing how to do this in the effects documentation. Help!


Thanks.



-- 
*Abraham Mathew
Statistical Analyst
www.amathew.com
720-648-0108
@abmathewks*

	[[alternative HTML version deleted]]

Why not use

?predict.glm ## with type = "response"  ?

-- Bert

On Mon, Aug 13, 2012 at 12:39 PM, Abraham Mathew <abmathewks at gmail.com>
wrote:
> I'm trying to run a logit model and plot the probability curve for a
number
> of the important predictors. I'm trying to do this
> with the Effects package.
>
>
> df=data.frame(income=c(5,5,3,3,6,5),
>               won=c(0,0,1,1,1,0),
>               age=c(18,18,23,50,19,39),
>               home=c(0,0,1,0,0,1))
> str(df)
>
> md1 = glm(factor(won) ~ income + age + home,
>           data=df, family=binomial(link="logit"))
>
> summary(md1)
>
> plot(effect("income", md1), grid=TRUE)
>
>
> But I want to know how to plot a graph so that it shows the probability of
> won (response) based on income (or any of the other predictors).
>
> However, what I want to do is generate the same plot, with won don y axis
> and income on x axis, but the curves showing the probabilities for age and
> home.
>
>
> Not seeing how to do this in the effects documentation. Help!
>
>
> Thanks.
>
>
>
> --
> *Abraham Mathew
> Statistical Analyst
> www.amathew.com
> 720-648-0108
> @abmathewks*
>
>         [[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.


-- 

Bert Gunter
Genentech Nonclinical Biostatistics

Internal Contact Info:
Phone: 467-7374
Website:
http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm

With multiple predictors, I feel that predict ends up being less elegant.
Plus, I've found a lot of the
simple log-odds plots available in the effect package to be simple and easy
to use as compared to
base predict.

On Mon, Aug 13, 2012 at 3:01 PM, Bert Gunter <gunter.berton@gene.com>
wrote:
> Why not use
>
> ?predict.glm ## with type = "response"  ?
>
> -- Bert
>
> On Mon, Aug 13, 2012 at 12:39 PM, Abraham Mathew
<abmathewks@gmail.com>
> wrote:
> > I'm trying to run a logit model and plot the probability curve for
a
> number
> > of the important predictors. I'm trying to do this
> > with the Effects package.
> >
> >
> > df=data.frame(income=c(5,5,3,3,6,5),
> >               won=c(0,0,1,1,1,0),
> >               age=c(18,18,23,50,19,39),
> >               home=c(0,0,1,0,0,1))
> > str(df)
> >
> > md1 = glm(factor(won) ~ income + age + home,
> >           data=df, family=binomial(link="logit"))
> >
> > summary(md1)
> >
> > plot(effect("income", md1), grid=TRUE)
> >
> >
> > But I want to know how to plot a graph so that it shows the
probability
> of
> > won (response) based on income (or any of the other predictors).
> >
> > However, what I want to do is generate the same plot, with won don y
axis
> > and income on x axis, but the curves showing the probabilities for age
> and
> > home.
> >
> >
> > Not seeing how to do this in the effects documentation. Help!
> >
> >
> > Thanks.
> >
> >
> >
> > --
> > *Abraham Mathew
> > Statistical Analyst
> > www.amathew.com
> > 720-648-0108
> > @abmathewks*
> >
> >         [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > R-help@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.
>
>
>
> --
>
> Bert Gunter
> Genentech Nonclinical Biostatistics
>
> Internal Contact Info:
> Phone: 467-7374
> Website:
>
>
http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm
>


-- 
*Abraham Mathew
Statistical Analyst
www.amathew.com
720-648-0108
@abmathewks*

	[[alternative HTML version deleted]]

Dear Abraham,

It's not entirely clear to me what you want to plot. If you want the
"effect
display" for each predictor holding the others to typical values,
plot(allEffects(md1)) will do that. 

But you appear to want to plot fitted probabilities for combinations of the
predictors even though the model is additive. If that's really what you
want, you should be able to get it by specifying
plot(effect("income*age*home", md1). This will generate a warning
because
"income:age:home" isn't a high-order term in the model, but it
should work.

I hope this helps,
 John

-------------------------------------------------------
John Fox
Senator William McMaster Professor of Social Statistics
Department of Sociology
McMaster Univeristy
Hamilton, Ontario, Canada


> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at
r-project.org]
> On Behalf Of Abraham Mathew
> Sent: August-13-12 12:40 PM
> To: r-help at r-project.org
> Subject: [R] Using the effects package to plot logit probabilities
> 
> I'm trying to run a logit model and plot the probability curve for a
number of
> the important predictors. I'm trying to do this with the Effects
package.
> 
> 
> df=data.frame(income=c(5,5,3,3,6,5),
>               won=c(0,0,1,1,1,0),
>               age=c(18,18,23,50,19,39),
>               home=c(0,0,1,0,0,1))
> str(df)
> 
> md1 = glm(factor(won) ~ income + age + home,
>           data=df, family=binomial(link="logit"))
> 
> summary(md1)
> 
> plot(effect("income", md1), grid=TRUE)
> 
> 
> But I want to know how to plot a graph so that it shows the probability of
> won (response) based on income (or any of the other predictors).
> 
> However, what I want to do is generate the same plot, with won don y axis
> and income on x axis, but the curves showing the probabilities for age and
> home.
> 
> 
> Not seeing how to do this in the effects documentation. Help!
> 
> 
> Thanks.
> 
> 
> 
> --
> *Abraham Mathew
> Statistical Analyst
> www.amathew.com
> 720-648-0108
> @abmathewks*
> 
> 	[[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.