R help - Jun 2016 - ggplot2 stat_smooth

I want to add a logistic plot to data.

My call to ggplot is:


ggplot(data = SSI.dt, aes(x = elapsed, y = 1 - control)) + geom_point() +
stat_smooth(method = 'glm', family = binomial) +
xlab('Surgery Duration (min)') + ylab('Probability SSI') +
labs(title = 'THA Surgical Site Infections')
ggsave(filename = 'Plots/SSI.Duration.pdf?)

An error message is returned: Unknown parameters: family

Removing ?family = binomial? returns a straight line with points appropriately
placed on y = 0 and y = 1.

I found some previous messages on markmail that listed my call as the correct
syntax structure.

I?d appreciate thoughts/pointers.


Nathan

-- 
Nathan Pace, MD, MStat
Department of Anesthesiology
University of Utah
801.581.6393
n.l.pace at utah.edu

> On Jun 29, 2016, at 2:17 PM, Nathan Pace <n.l.pace at utah.edu>
wrote:
> 
> I want to add a logistic plot to data.
> 
> My call to ggplot is:
> 
> 
> ggplot(data = SSI.dt, aes(x = elapsed, y = 1 - control)) + geom_point() +
> stat_smooth(method = 'glm', family = binomial) +
> xlab('Surgery Duration (min)') + ylab('Probability SSI') +
> labs(title = 'THA Surgical Site Infections')
> ggsave(filename = 'Plots/SSI.Duration.pdf?)
> 
> An error message is returned: Unknown parameters: family
So you would have naturally looked at the acceptable arguments for the function,
right?
> 
> Removing ?family = binomial? returns a straight line with points
appropriately placed on y = 0 and y = 1.
> 
> I found some previous messages on markmail that listed my call as the
correct syntax structure.
Perhaps they referred to older versions of the function.

Looking at:

?stat_smooth

 The current help page implements this by creating a helper function,
binomial_smooth,  but using its example of the needed arguments I did just try:

... + stat_smooth(method = 'glm', method.args = list(family =
"binomial"))


With success.

-- 
David.
> 
> I?d appreciate thoughts/pointers.
> 
> 
> Nathan
> 
> -- 
> Nathan Pace, MD, MStat
> Department of Anesthesiology
> University of Utah
> 801.581.6393
> n.l.pace at utah.edu
> 
> 
> 
> 
> 
> 
> 
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.
David Winsemius
Alameda, CA, USA

I appreciate the pointers.

I am using ggplot2 2.1.0, but I was looking at the wrong version of the web page
example documentation.

The easiest way is to define the function

binomial_smooth <- function(...) {
  geom_smooth(method = "glm", method.args = list(family =
"binomial"), ...)
}



And then call

ggplot(data = SSI.dt, aes(x = elapsed, y = 1 - control)) + geom_point() +

binomial_smooth()

Everything is now working smoothly.


Nathan
 





-----Original Message-----
From: David Winsemius <dwinsemius at comcast.net>
Date: Wednesday, June 29, 2016 at 16:10
To: Nathan L Pace <n.l.pace at utah.edu>
Cc: "r-help at r-project.org" <r-help at r-project.org>
Subject: Re: [R] ggplot2 stat_smooth
>
>> On Jun 29, 2016, at 2:17 PM, Nathan Pace <n.l.pace at utah.edu>
wrote:
>> 
>> I want to add a logistic plot to data.
>> 
>> My call to ggplot is:
>> 
>> 
>> ggplot(data = SSI.dt, aes(x = elapsed, y = 1 - control)) + geom_point()
+
>> stat_smooth(method = 'glm', family = binomial) +
>> xlab('Surgery Duration (min)') + ylab('Probability
SSI') +
>> labs(title = 'THA Surgical Site Infections')
>> ggsave(filename = 'Plots/SSI.Duration.pdf?)
>> 
>> An error message is returned: Unknown parameters: family
>
>So you would have naturally looked at the acceptable arguments for the
function, right?
>
>> 
>> Removing ?family = binomial? returns a straight line with points
appropriately placed on y = 0 and y = 1.
>> 
>> I found some previous messages on markmail that listed my call as the
correct syntax structure.
>
>Perhaps they referred to older versions of the function.
>
>Looking at:
>
>?stat_smooth
>
> The current help page implements this by creating a helper function,
binomial_smooth,  but using its example of the needed arguments I did just try:
>
>... + stat_smooth(method = 'glm', method.args = list(family =
"binomial"))
>
>
>With success.
>
>-- 
>David.
>> 
>> I?d appreciate thoughts/pointers.
>> 
>> 
>> Nathan
>> 
>> -- 
>> Nathan Pace, MD, MStat
>> Department of Anesthesiology
>> University of Utah
>> 801.581.6393
>> n.l.pace at utah.edu
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> ______________________________________________
>> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
>> 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.
>
>David Winsemius
>Alameda, CA, USA
>