plot pch

Thankyou very much rui.i was trying the following command but i was unsuccessful in implementing the secondary axis.

library(lattice)> df <- data.frame(x = x, y = y, z= z)> > xyplot(x~y,data=df,groups=z,+        par.settings=list(superpose.symbol=list(pch=1:4,+                                                col='black')))

By the way, is there a way that instead of making a separate secondary axis, we could put a small classification number on the top of every symbol?? Thankyou very much for helping me out....
Elisa

Date: Mon, 11 Mar 2013 21:33:33 +0000
From: ruipbarradas at sapo.pt
To: eliza_botto at hotmail.com
CC: r-help at r-project.org
Subject: Re: [R] plot pch

Hello,

Try the following.

dat <-
structure(list(x = c(52.5211966906669, 49.8818953734405, 45.176643179385,
52.2988909042952, 46.1007828702965, 53.4910266471472, 49.013651970548,
55.1218462810432, 56.922347550005, 50.7531277271902), y =
c(23.953008533893,
45.1590889001506, 62.5358474328017, 39.4861634272874, 49.3108455233261,
46.0010163238786, 46.5779093951996, 37.4552698003145, 46.4057156535496,
56.667549695452), z = c(1, 4, 1, 2, 1, 3, 4, 2, 1, 3), r = 1:10), .Names
= c("x",
"y", "z", "r"), row.names = c(NA, -10L), class = "data.frame")

ch <- c(2, 22, 5, 1, 4)  # triangle, rectangle, rhombus, circle, cross.
plot(dat$x, dat$y, pch = ch[dat$z])
axis(4, at = dat$y, labels = seq_along(dat$y)[order(dat$y)])


Also, if you want to see the pch characters, run without worrying about
the warnings (in general a bad tip, not really important here)

plot(1, type = "n", xlim=c(0, 17), ylim=c(0, 17))
for(x in 1:16) for(y in 1:16)
	points(x, y, pch = (x - 1)*16 + y)


They go first column with x = 1, from y = 1 to 16, etc.


Hope this helps,

Rui Barradas

Em 11-03-2013 17:43, eliza botto escreveu:

Dear xpeRts,

I would like to ask a question about plotting in R.
I have four columns in which first column contains x-coordinate and second column contains y-coordinate of certain points, whereas the third column "z" contains classification number of those points from 1 to 4, it should be noted that this column doesn't quantify anything, but just classify the points. More precisely like the following

dput(v)

structure(list(x = c(52.5211966906669, 49.8818953734405, 45.176643179385,
52.2988909042952, 46.1007828702965, 53.4910266471472, 49.013651970548,
55.1218462810432, 56.922347550005, 50.7531277271902), y = c(23.953008533893,
45.1590889001506, 62.5358474328017, 39.4861634272874, 49.3108455233261,
46.0010163238786, 46.5779093951996, 37.4552698003145, 46.4057156535496,
56.667549695452), z = c(1, 4, 1, 2, 1, 3, 4, 2, 1, 3), r = 1:10), .Names = c("x",
"y", "z", "r"), row.names = c(NA, -10L), class = "data.frame")


What i want to do is to locate, in the space of "x" and "y", each point (in classified section) with different shape for each classified section range from 1 to 4, by command "pch".
There should also be a y-axis on the right end of x-axis (secondary axis), which carries the the point number as in column "r". As there are ten points so that axis should range to 10 for the presentation of each point number. More precisely, z=1 should be located at a point where x=52.52120, y=23.95301 and r=1.
where,
x= x-axis

y=y-axis on the left end of x-axis

r=y-axis on the right end of x-axis (secondary axis)

For points belonging to classification 1 there should be a triangle
For points belonging to classification 2 there should be a rectangle
and so on...
in my orinigal data i have 5 classifications.

i hope i was clear, but in case i my wording is confusing, kindly let me know.

Elisa
   		 	   		
	[[alternative HTML version deleted]]

  		 	   		
	[[alternative HTML version deleted]]