to Introduction R
Downloading
For your home/office computer, you can get binaries (ready-to-run) for R by going to http://www.r-project.org/ clicking on CRAN and picking one of "mirror" websites to download from, click on Windows, base, R-2.2.1-win32.exe and then run it. In the TCC labs, R is already installed.
Getting help
If you know the name of the topic, you can type help(topic) in the R console. I often prefer to use browsable help available at file:///C:/Program%20Files/R/rw2101/doc/html/packages.html that can also be recalled by typing in help.start() There are also PDF manuals available for most packages.
Intro
Both R and Matlab are based on similar programming principles. Both are casesensitive. A distinct feature of R (partly, Matlab, too) is that it's built out of packages. The basic packages we'll need are basic, graphics and stats. Additional packages with geostat functions are geoR and RandomFields. Most packages are user-contributed (buyer beware!). To load a package, type in library(package_name). See the document http://infohost.nmt.edu/~olegm/586/InstallingPackages.rtf for specific instructions. The R program can be typed in line by line (use the up arrow to recall previously entered lines), but I usually type it in some editor (e.g. Notepad) and then paste the contents into the R console. You are welcome to look at the programs used to generate the Lecture notes output at http://infohost.nmt.edu/~olegm/586/Rcode/
1
Comparative table
R x <- 2 # X if you want printing ... { ; ... ... } Matlab x=2 % X; if you don't want printing , ... end plotting plot(x,y) plot(x,y, type="l") contour(X,Y,Z) scatter(x,y) plot(x,y) contour(X,Y,Z) matrix/vector operations c(1,2,3) rbind(c(1,2,3),c(4,5,6)) A <- matrix(0, n1, n2) dim(A) x+y, x*y, x^2, sin(x) ... t(x) A %*% B A[1:5,2:4] A[1:5,] solve(A,b) 1:n seq(a,b,by = step) X <- read.table("U: /Data.dat") write.table(X,....) abs, sin, exp, log, ... rnorm(n,mean,var) runif(n,a,b) mean(x), sd(x), var(x) cor(x,y), cov(x,y) if (q==1) x <- 5 else x <- 2 for (i in 1:n){ ... } f <- function(x){ ... out} [1 2 3] [ 1 2 3; 4 5 6] A = zeros(n1,n2) size(A) x+y, x.*y, x.^2, sin(x) x' A*B A(1:5,2:4) A(1:5,:) b\A 1:n a:step:b interface load Data.dat; (reads into a matrix with same name) save Data.dat X -ascii some functions abs, sin, exp, log, ... normrnd (MU,SIGMA) unifrnd(A,B) mean(x), std(x), var(x) corrco...