a problem with crossbar (moving to ggplot0.9)

Domenico Vistocco

unread,

May 3, 2012, 10:33:22 AM5/3/12

to ggp...@googlegroups.com

Dear All,

I am struggling with the ggplot2_0.9 in order to reuse my previous code. I asked Had for an help and he suggests me to post again my request to the list (trying to provide a reproducible example). Here it is.

Sorry for posting again the question, I try to better explain my problem.

The following code worked perfectly for me on a linux system with ggplot2_0.8.5 on an older R version:

--------------------------------------------------------------------------------------

library(ggplot2)

library(MASS)

stat_sum_single <- function(fun, geom="point", colour="red", size=3, ...) {

stat_summary(fun.y=fun, colour=colour, geom=geom, size=size, ...)

}

stat_sum_crossbar <- function(fun1, fun2, fun3, geom="crossbar", colour="blue", ...) {

stat_summary(fun.y=fun1, fun.ymin=fun2, fun.ymax=fun3, colour=colour, geom=geom, ...)

}

carsRidotto <- subset(Cars93, Passengers %in% 4:6)

ggp <- qplot(data= carsRidotto, Passengers, Price, geom="point")

ggp2 <- ggp + stat_sum_single(mean, shape=8, size=8)

ggp2 <- ggp2 + stat_sum_single(function(x) quantile(x, 0.1), colour="orange", shape=17, size=7) +

stat_sum_single(function(x) quantile(x, 0.9), colour="orange", shape=17, size=7)

ggp2 <- ggp2 + stat_sum_crossbar(function(x) quantile(x, 0.5),

function(x) quantile(x, 0.25),

function(x) quantile(x, 0.75),

geom="crossbar", colour="blue", width=0.5)

ggp2

ggp3 <- ggp2 +

stat_quantile(colour="blue", size=1, linetype=2) +

stat_quantile(colour="orange", size=1, linetype="twodash",quantiles=c(0.1, 0.9)) +

stat_smooth(method=lm, se=FALSE, colour="red")

ggp3

--------------------------------------------------------------------------------

Here you can find the three resulting graphs:

http://stat.unicas.it/downloadStatUnicas/appoggio/fig-ggp.pdf

http://stat.unicas.it/downloadStatUnicas/appoggio/fig-ggp2.pdf

http://stat.unicas.it/downloadStatUnicas/appoggio/fig-ggp3.pdf

and here is my sessionInfo():

http://stat.unicas.it/downloadStatUnicas/appoggio/sessionInfo.txt

Using the last version of ggplot2 on a Mac OsX installation, I have problem with the second plot. Here is my new sessionInfo()

http://stat.unicas.it/downloadStatUnicas/appoggio/sessionInfoNew.txt

In particular here is the resulting "bad" plot:

http://stat.unicas.it/downloadStatUnicas/appoggio/bad-fig-ggp2.pdf

The third plot correctly adds the regression lines to the second plot (I think the problem lies in the stat_sum_crossbar function). Here is the "bad" third plot:

http://stat.unicas.it/downloadStatUnicas/appoggio/bad-fig-ggp3.pdf

Thanks to the help of Brian Diggs, I succeed in reproduce the ggp2 plot using the group aesthetic in the qplot, i.e. as follows:

ggp <- qplot(data= carsRidotto, Passengers, Price, geom="point", group=Passengers)

but using such change I am not more able to reproduce the third plot.

I really appreciate any suggestion.

Ciao,

domenico

adam.l...@pnc.com

unread,

May 3, 2012, 11:17:39 AM5/3/12

to Domenico Vistocco, ggp...@googlegroups.com

Hi Domenico,

This works. The group aesthetic fixes it. This works:

ggplot(carsRidotto,aes(Passengers,Price))+
geom_point()+
stat_summary(fun.y ="mean",colour="red",size=4,shape=8,geom="point")+
stat_summary(fun.y = function(x) quantile(x,.9), colour="orange", shape=17, size=7, geom="point")+
stat_summary(fun.y = function(x) quantile(x,.1), colour="orange", shape=17, size=7, geom="point")+

stat_summary(aes(Passengers,Price,group=Passengers),fun.y = function(x) quantile(x,.5), fun.ymax = function(x) quantile(x,.75)
,fun.ymin = function(x) quantile(x,.25)
,width=0.5,geom="crossbar",colour="blue")+

stat_quantile(colour="blue", size=1, linetype=2) +
stat_quantile(colour="orange", size=1, linetype="twodash",quantiles=c(0.1, 0.9)) +
stat_smooth(method=lm, se=FALSE, colour="red")

Adam Loveland

From:	Domenico Vistocco <domenico...@gmail.com>
To:	ggp...@googlegroups.com
Date:	05/03/2012 10:33 AM
Subject:	a problem with crossbar (moving to ggplot0.9)
Sent by:	ggp...@googlegroups.com

--
You received this message because you are subscribed to the ggplot2 mailing list.
Please provide a reproducible example: https://github.com/hadley/devtools/wiki/Reproducibility

To post: email ggp...@googlegroups.com
To unsubscribe: email ggplot2+u...@googlegroups.com
More options: http://groups.google.com/group/ggplot2

The contents of this email are the property of PNC. If it was not addressed to you, you have no legal right to read it. If you think you received it in error, please notify the sender. Do not forward or copy without permission of the sender. This message may contain an advertisement of a product or service and thus may constitute a commercial electronic mail message under US Law. The postal address for PNC is 249 Fifth Avenue, Pittsburgh, PA 15222. If you do not wish to receive any additional advertising or promotional messages from PNC at this e-mail address, click here to unsubscribe. https://pnc.p.delivery.net/m/u/pnc/uni/p.asp
By unsubscribing to this message, you will be unsubscribed from all advertising or promotional messages from PNC. Removing your e-mail address from this mailing list will not affect your subscription to alerts, e-newsletters or account servicing e-mails.

Domenico Vistocco

unread,

May 3, 2012, 1:35:48 PM5/3/12

to ggp...@googlegroups.com, Domenico Vistocco

Dear Alan,

thanks for the suggestions: I used the group aesthetic in the qplot call and not in the stat_summary call.

I will try to adapt my code using your tip.

Thanks a lot,

domenico

To post: email ggp...@googlegroups.com
To unsubscribe: email ggplot2+unsubscribe@googlegroups.com
More options: http://groups.google.com/group/ggplot2

Reply all

Reply to author

Forward