R-Cookbook.com Delicious, statistical-computing recipes

The R plotting symbols pch=21:25 have borders and backgrounds. col gives the border color and bg the background color. (R version 2.8.x)

Example:

opar<-par(mar=c(0,0,2,0),oma=c(0,0,0,0))
plot(1:5,rep(1,5),col='blue',bg='red',pch=21:25,cex=4,
   lwd=2,axes=FALSE,xlim=c(0,6),ylim=c(0.5,1.5),
   main="two-colored symbols",xlab="")
text(1:5,rep(1,5),labels="21":"25")
par(opar)

# Another version using location and substrings:

x <- as.POSIXct("2005-09-01 01:25:46.9")
print(x) #note R didn't round up seconds, and added timezone

yr <- substring(x,1,4)
m <- substring(x,6,7)
d <- substring(x,9,10)
hour <- substring(x,12,13)
min <- substring(x,15,16)
sec <- substring(x,18,19) #note, I decided to truncate decimals,
#you don't have to

## This is simple and you only have to print a variable first, to see
## how it is arranged.

Our data in multivariate (wide) format:

my_data<- data.frame(
    id=c(1:50),
    depression1=rnorm(50),
    anxiety1=rnorm(50),
    depression2=rnorm(50),
    anxiety2=rnorm(50),
    depression3=rnorm(50),
    anxiety3=rnorm(50)
)
my_data

Read more after the jump.

Read more

This function will extract time labeled variables (or any variables with consistent naming) by name from a data frame in the order in which they appear in the data frame. This is especially useful in cases when one has a longitudinal or time-series data set where each row (subject) has many occasions of measurements for each measure. Code and examples after the jump.

Read more

Forgot to post this:

https://stat.ethz.ch/pipermail/r-announce/2008/000841.html

The input is specified as the row-wise lower triangle or column-wise upper triangle.

Read more

This is an overview of RPy:

http://www.daimi.au.dk/~besen/TBiB2007/lecture-notes/rpy.html

While I've never used it, I can definitely see doing so, as I've recently become more and more a python programmer.

The function and documentation is listed after the jump.

Read more

Creating a PDF version of your graphs is very easy:

pdfname<-paste("myfilename",".pdf",sep="")
pdf(pdfname, height=6.4,width=6.4)
# Plotting code
dev.off()

From the r-help mailing list:

x <- rnorm(10)
y <- runif(10)
z <- 1:10

plot(x~z, ylim=range(x, y), type='l')
lines(y, col='red')