• Execute Installation of R • Manipulate Data With R • Review Programming What is R • R is a free software for statistical computation and graphics
• Operation Systems: Windows, Mac, or Linux
• Homepage: http://www.r-project.org Installing R Under Windows Suppose you have a laptop with Windows OS(32/64 bits), how to install R in your laptop? • Go to any CRAN site (see http://cran.r-project.org for a list), and click “Download R for Windows.” • Follow the instruction to click ‘install R for the first time” • Download R installation file, double-click on the icon and follow the instructions to install. As of January 2020, the file is “R-3.6.2-win.exe” (Size: 83Mb). Data With R • Objects: vector, matrix, data.frame, ts, list, factor, array • Mode (numerical, character, complex, and logical);Length • Read data stored in text (ASCII) files read.table(), scan(), and read.fwf() • Saving data write(x, file=“data.txt”), write.table() write in a file a data.frame Generating Data in R # Generating data x1 <- 1:10 # x1 = c(1,2,3,4,5,6,7,8,9,10) x2 <- seq(1,4,0.5) # x2 = c(1.0,1.5, 2.0, 2.5, 3.0, 3.5, 4.0) x3 <- rep(1,5) # x3 = c(1,1,1,1,1) x4 <- gl(2,3,10) (generate factor levels)
#Check x4 values in R > x4 [1] 1 1 1 2 2 2 1 1 1 2 Levels: 1 2 Matrix in R # matrix in R y <- matrix(c(1,2,3,4,5,6),2,3) # This is 2x3 Matrix filled in column by column [,1] [,2] [,3] [1,] 1 3 5 [2,] 2 4 6
w <- matrix(c(1,2),2,1) # This is 2x1 Matrix/vector
t(w) %*% y #matrix product w^T y= c(5, 11, 17) Matrix Operations in R (cont.) # matrix in R y <- matrix(c(1,2,3,4,5,6),2,3)
y[1,3] # is the element on row 1 and column 3
y[1,] # is the first row, c(1,3,5) y[,2] # is the second column, c(3,4) apply(y,2,mean) #gives us the column means, c(1.5, 3.5, 5.5) apply(y,1,sd) #gives us the standard deviation of each row, c(2, 2) Data Frame #Generate 100 normal variables x <- rnorm(100) # w will be used as a `weight’ vector w <- 1 + x/2 # Model to generate Y from x and w y <- x + w * rnorm(x) # Make a data frame of three columns named x,y, w, and look at it. dum <- data.frame(x,y,w) dum Linear Regression in R # Fit a simple ordinary linear regression of y and x fm1 <- lm(y ~ x, data= dum); summary(fm1)
# we do a weighted least squares regression
fm2 <- lm( y ~ x, data = dum, weight = 1/ w^2) summary(fm2)
# Plots plot(x,y); abline(fm1); abline(fm2, col=“red”) Programming Language • You can write your own function in R, and then call that function later. • R is rather slow in loops, and you should avoid loops as much as possible by using matrix operations.
Example: suppose we have a vector x=(-2,3,5,-8), and for each element of x
with the value 3, we want to give the value of 0 to another variable y, else 1. • R code: x <- c(-2,3,5,-8); y <- (x != 3);
• R output: y [1] TRUE FALSE TRUE TRUE
More R • R studio: An integrated development environment for R. https://www.rstudio.com
• Books related to R: https://www.r-project.org/doc/bib/R-books.html
