> On Sep 25, 2016, at 6:43 PM, John Sorkin wrote:
>
> I have a data frame that contains data for multiple (seven) subjects. Each
> subject is represented by a new value of PID.
> I would like to plot the data for all seven subjects. For each subject I want
> to plot a line showing CT as a func
Hi John,
I know this is kind of dumb, but:
plot(0,xlim=range(xx$Nit,na.rm=TRUE),
ylim=range(xx$CT,na.rm=TRUE),type="n",
xlab="Nit",ylab="CT")
for(i in unique(xx$PID))
points(xx$Nit[xx$PID==i],xx$CT[xx$PID==i],
pch=i,col=i,type="b")
Jim
On Mon, Sep 26, 2016 at 11:43 AM, John Sorkin
wrote:
>
Hi Richard,
Thanks for demonstrating the usage of layer in latticeExtra. In my current
case I do have multiple panels on one page but only need some universal
annotation text for the whole page. So "sub" argument does the job pretty
well. You have certainly shown a very interesting example to add
I have a data frame that contains data for multiple (seven) subjects. Each
subject is represented by a new value of PID.
I would like to plot the data for all seven subjects. For each subject I want
to plot a line showing CT as a function of Nit, with the dots for each subject
joined. I have tri
Hi Bert,
?? Dont you already have the answer: you must set warn.conflicts =
FALSE in all your library calls.
Thanks. This works. Now understand the logic but first I was thinking
that removing WARNING messages should remove conflit warning as well.
Anyway it works.
Thanks
Cheers
Fabien
Ch
Bert,
I interpreted the OP's request as a single panel xyplot.
The latticeExtra::layer mechanism does work for multiple plots.
library(lattice)
library(latticeExtra)
tmp <- data.frame(x=1:10, y=1:10, g=rep(c("A","B"), each=5))
xyplot(y ~ x | g, data=tmp)
xyplot(y ~ x | g, data=tmp, main="outsi
?? Dont you already have the answer: you must set warn.conflicts =
FALSE in all your library calls.
Cheers,
Bert
Bert Gunter
"The trouble with having an open mind is that people keep coming along
and sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
Try enclosing the whole thing in "suppressMessages";
suppressMessages({
if (!require("plyr",quietly=TRUE,warn.conflicts=FALSE))
suppressMessages(install.packages("plyr",quietly=TRUE))
if (!require("dplyr",quietly=TRUE))
suppressMessages(install.packages("dplyr",quietly=TRUE))
if (!requ
Presumably the "earth" package lacks this functionality ...?
So, obvious query: did you try using the boot package? If not, why
not? If so, show us the code that failed.
Or am I missing the point?
Cheers,
Bert
Bert Gunter
"The trouble with having an open mind is that people keep coming along
... but what if there are multiple panels?
"The layer mechanism is a method for augmenting a panel function. It
allows expressions to be added to the panel function without knowing
what the original panel function was. "
As I understand it, the OP requested annotation for the entire lattice
displ
Hi there,
I would like to remove all messages when I load library (I fact I am
using "require" and "install.packages"). I tried many options and look
at the documentation for the 2 functions.
For example I am using the following piece of code: init.R
print("step 1")
# Install library if not i
Dear R-experts,
I have fitted a MARS regression and am trying now to plot/draw the BCa
confidence bands around the 3 fitted curves (QUALITESANSREDONDANC,
competitivite and innovation).
Here is the reproducible example.
Dataset =
data.frame(PIBparHab=c(43931,675
library(lattice)
library(latticeExtra)
tmp <- data.frame(x=1:10, y=1:10)
xyplot(y ~ x, data=tmp)
xyplot(y ~ x, data=tmp, main="text outside panel clipped") +
## "+" must be on same line as first statement.
## This use of "+" is from latticeExtra.
latticeExtra::layer(panel.text(x=2, y=-0.5,
This illustrates why you need to post a reproducible example. You have a
number of confounding factors in your code.
First, "data" is a commonly-used function... avoid using it for variable
names.
Second, using the attach function this way leads to confusion... best to
forget this function u
> On 25 Sep 2016, at 19:37, Matti Viljamaa wrote:
>
> Okay here’s a pretty short code to reproduce it:
>
> data <-
> read.table("http://users.jyu.fi/~slahola/files/glm1_datoja/yoruotsi.txt";,
> header=TRUE)
data$clka <- I(data$lka - mean(data$lka))
> attach(data)
>
> fit2 <- lm(ruotsi.pist
Object clka not found.
Did you test run it in a fresh R environment?
--
Sent from my phone. Please excuse my brevity.
On September 25, 2016 9:37:56 AM PDT, Matti Viljamaa wrote:
>Okay here’s a pretty short code to reproduce it:
>
>data <-
>read.table("http://users.jyu.fi/~slahola/files/glm1_da
Hello Dr. Fox,
Thank you very much for your recommendations and I will take action on
following them. I appreciate your willingness to provide guidance on my novice
questions.
Sincerely,
Courtney
From: Fox, John
Sent: Sunday, September 25, 2016 9:21 AM
Okay here’s a pretty short code to reproduce it:
data <-
read.table("http://users.jyu.fi/~slahola/files/glm1_datoja/yoruotsi.txt";,
header=TRUE)
attach(data)
fit2 <- lm(ruotsi.pist ~ mies + koulu + lka + koulu*clka)
bs <- coef(fit2)
varitB <- c(data[koulu == 'B',]$mies)
varitB[varitB == 0] =
Thanks Bert, sub works pretty well.
Carlos, thanks for the suggestion. As I am not familiar with latticeExatra,
not sure how hard it is to make it work with my plot. I'll try.
On Sat, Sep 24, 2016 at 11:04 AM, Bert Gunter
wrote:
> No.
>
> ltext/panel.text plots text *within* panels; IIUC, the
Go directly to C. Do not pass go, do not collect $200.
You think curve does something, but you are missing what it actually does.
Since you don't seem to be learning from reading ?curve or from our responses,
you need to give us an example you can learn from.
--
Sent from my phone. Please excu
On 2016-09-25 18:52, Jeff Newmiller wrote:
You seem to be confused about what curve is doing vs. what you are
doing.
But my x-range in curve()'s parameters from and to should be the entire
lka vector, since they are from=min(lka) and to=max(lka). Then why does
this not span the entire of lka?
You seem to be confused about what curve is doing vs. what you are doing.
A) Compute the points you want to plot and put them into 2 vectors. Then figure
out how to plot those vectors. Then (perhaps) consider putting that all into
one line of code again.
B) The predict function is the preferr
On 2016-09-25 18:30, Duncan Murdoch wrote:
On 25/09/2016 9:10 AM, Matti Viljamaa wrote:
Writing:
bs["(Intercept)"]+bs["mies"]*0+bs["kouluB"]+bs["lka"]*lka+bs["kouluB:clka"]*clka
i.e. without that being inside curve produces a vector of length 375.
So now it seems that curve() is really skippi
On 25/09/2016 7:51 AM, Matti Viljamaa wrote:
I have created a 2x2 plot using par(mfrow = c(2, 2)).
I can add x- and ylabels to individual plots, but what I want is to add overall
xlabel and ylabel for the entire 2x2 plot.
How to do this?
First you need to make space using par(oma=...), then
On 25/09/2016 9:10 AM, Matti Viljamaa wrote:
Writing:
bs["(Intercept)"]+bs["mies"]*0+bs["kouluB"]+bs["lka"]*lka+bs["kouluB:clka"]*clka
i.e. without that being inside curve produces a vector of length 375.
So now it seems that curve() is really skipping some lka-/x-values.
How could curve() k
Dear Courtney,
You're confusing a function call, na.action(na.omit), with an argument
specification, na.action=na.omit (and, in any event, there is no na.action()
function). But you don't have to specify na.action=na.omit, because na.omit
(which produces a complete-case analysis) is the defaul
Writing:
bs["(Intercept)"]+bs["mies"]*0+bs["kouluB"]+bs["lka"]*lka+bs["kouluB:clka"]*clka
i.e. without that being inside curve produces a vector of length 375.
So now it seems that curve() is really skipping some lka-/x-values.
> On 25 Sep 2016, at 16:01, Matti Viljamaa wrote:
>
> I’m trying
Hi,
Here's a place to start...
https://www.r-bloggers.com/two-tips-adding-title-for-graph-with-multiple-plots-add-significance-asterix-onto-a-boxplot/
You might also want to checkout the text() function and the xpd argument to par.
Ben
> On Sep 25, 2016, at 7:51 AM, Matti Viljamaa wrote:
>
>
I’m trying to plot regression lines using curve()
The way I do it is:
bs <- coef(fit2)
and then for example:
curve(bs["(Intercept)"]+bs["mies"]*0+bs["kouluB"]+bs["lka"]*x+bs["kouluB:clka"]*clka,
from=min(lka), to=max(lka), add=TRUE, col='red')
This above code runs into error:
Error in curve(
I have created a 2x2 plot using par(mfrow = c(2, 2)).
I can add x- and ylabels to individual plots, but what I want is to add overall
xlabel and ylabel for the entire 2x2 plot.
How to do this?
__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and
30 matches
Mail list logo
