Scatter plot / LOESS, or LOWESS for more than one parameter

Hi folks.

If I have the following in my "data"

event ? ?pH1 ? ?pH2
1 ? ? ? ? ? ?4.0 ? ? 6.0
2 ? ? ? ? ? ?4.3 ? ? 5.9
3 ? ? ? ? ? ?4.1 ? ? 6.1
4 ? ? ? ? ? ?4.0 ? ? 5.9
and on and on..... for about 400 events

Is there a way I can get R to plot event vs. pH1 ?and event vs. pH2 and
then do a loess or lowess line for each??

Thanks in advance
David

? ? ? ?[[alternative HTML version deleted]]

The scatter plot is easy:

plot(pH1 ~ pH2, data = OBJ)

When you say a loess for each -- how do you break them up? Are there
repeat values for pH1? If so, this might be hard to do in base
graphics, but ggplot2 would make it easy:

library(ggplot2)
ggplot(OBJ, aes(x = pH1, y = pH2)) + geom_point() + stat_smooth() +
facet_wrap(~factor(pH1))

or something similar.

Michael

On Mon, Apr 23, 2012 at 11:26 PM, David Doyle <kydaviddoyle at gmail.com> wrote:

Hi folks.

If I have the following in my "data"

event ? ?pH1 ? ?pH2
1 ? ? ? ? ? ?4.0 ? ? 6.0
2 ? ? ? ? ? ?4.3 ? ? 5.9
3 ? ? ? ? ? ?4.1 ? ? 6.1
4 ? ? ? ? ? ?4.0 ? ? 5.9
and on and on..... for about 400 events

Is there a way I can get R to plot event vs. pH1 ?and event vs. pH2 and
then do a loess or lowess line for each??

Thanks in advance
David

? ? ? ?[[alternative HTML version deleted]]

Hi Greg,

Sloved my own problem.

I had some missing data "NA" in the datasets. ?So I manually entered the
ylim=range(4,6)
and it worked!!!

Thanks!!
David


On Tue, Apr 24, 2012 at 1:55 PM, David Doyle <kydaviddoyle at gmail.com> wrote:

Hi Greg,

Thanks,

I got the 1st example to work using the following code:


data <- read.csv("http://doylesdartden.com/Monthly-pH-example.csv",
sep=",")

attach(data)

par(mfrow=c(2,1))
scatter.smooth( Year, MC.pH )
scatter.smooth( Year, MV.pH )


This is good but what I'm really looking for is to have them on the same
graph.


I tried your second example using the code below but got:

"Error in plot.window(...) : need finite 'ylim' values"

here is the code I used


data <- read.csv("http://doylesdartden.com/Monthly-pH-example.csv",
sep=",")

attach(data)


plot( Year, MC.pH, ylim=range(MC.pH,MV.pH) , col='blue')

points( Year, MV.pH, col='green' )

lines( loess.smooth(Year,MC.pH), col='blue')

lines( loess.smooth(Year,MV.pH), col='green')


Thanks again

David



On Tue, Apr 24, 2012 at 1:45 PM, Greg Snow <538280 at gmail.com> wrote:

Assuming that you want event as the x-axis (horizontal) you can do
something like (untested without reproducible data):

par(mfrow=c(2,1))
scatter.smooth( event, pH1 )
scatter.smooth( event, pH2 )

or

plot( event, pH1, ylim=range(pH1,pH2) , col='blue')
points( event, pH2, col='green' )
lines( loess.smooth(event,pH1), col='blue')
lines( loess.smooth(event,pH2), col='green')

Only do the second one if pH1 and pH2 are measured on the same scale
in a way that the comparison and any crossings are meaningful or if
there is enough separation (but not too much) that there is no
overlap, but still enough detail.



On Mon, Apr 23, 2012 at 10:40 PM, R. Michael Weylandt
<michael.weylandt at gmail.com> wrote:

The scatter plot is easy:

plot(pH1 ~ pH2, data = OBJ)

When you say a loess for each -- how do you break them up? Are there
repeat values for pH1? If so, this might be hard to do in base
graphics, but ggplot2 would make it easy:

library(ggplot2)
ggplot(OBJ, aes(x = pH1, y = pH2)) + geom_point() + stat_smooth() +
facet_wrap(~factor(pH1))

or something similar.

Michael

On Mon, Apr 23, 2012 at 11:26 PM, David Doyle <kydaviddoyle at gmail.com>
wrote:

Hi folks.

If I have the following in my "data"

event ? ?pH1 ? ?pH2
1 ? ? ? ? ? ?4.0 ? ? 6.0
2 ? ? ? ? ? ?4.3 ? ? 5.9
3 ? ? ? ? ? ?4.1 ? ? 6.1
4 ? ? ? ? ? ?4.0 ? ? 5.9
and on and on..... for about 400 events

Is there a way I can get R to plot event vs. pH1 ?and event vs. pH2
and
then do a loess or lowess line for each??

Thanks in advance
David

? ? ? ?[[alternative HTML version deleted]]

--
Gregory (Greg) L. Snow Ph.D.
538280 at gmail.com