R Studio

In R Studio there are (by default) four “panes” to work in.

These can all be configured, but we’ll stick with the default configuration for now.

Source Pane

The Source Pane provides several helpful features:

Console Pane

The console “is” R. You feed commands into it, and it produces output. It also keeps track of variables, functions, etc. that you define, and the previous commands you’ve given.

The console pane is where you directly interact with the console. You can do this by typing commands in directly, or by running them from the source pane.

The output from the console is often given in the console. Plots appear separately (by default in the “Plots/Help” pane).

Workspace

You can view a few different things in the upper right. By default you’ll see the Environment, where R shows you what things it’s keeping track of (Data frames, variables, functions). Many of these things, e.g. data frames and functions, can be opened in the source pane for viewing.

You can also view the commands the console has run, in the order it ran them.

Plots/Help

Plots is self-explanatory. Note that you can flip back and forth between various plots you’ve made.

The Help browser allows you to search for documentation on functions you are using, or might use. You can also access these from the console. Great when you know roughly what you want to do, but need to check syntax. Also, check the “See Also” at the bottom if you’re not sure what function you’re looking for.

Data Frames

A data frame is a fancy list. Let’s load the data frame from the lecture:

library(datasets)
states = data.frame(state.x77, Abbr=state.abb, 
                    Region=state.region, Division=state.division)
class(states)
## [1] "data.frame"
typeof(states)
## [1] "list"

Basic Access: Row

states['Alabama',]
##         Population Income Illiteracy Life.Exp Murder HS.Grad Frost  Area
## Alabama       3615   3624        2.1    69.05   15.1    41.3    20 50708
##         Abbr Region           Division
## Alabama   AL  South East South Central

Basic Access: Column

head( states['Population'] ,10)
##             Population
## Alabama           3615
## Alaska             365
## Arizona           2212
## Arkansas          2110
## California       21198
## Colorado          2541
## Connecticut       3100
## Delaware           579
## Florida           8277
## Georgia           4931

Also states$Population

Basic Access: Single Entry

states['Alabama','Population']
## [1] 3615

Fancy Access: Slice

“Slicing” a data frame returns a new data frame with the desired rows and/or columns

states[1:4,c('Population','Income')]
##          Population Income
## Alabama        3615   3624
## Alaska          365   6315
## Arizona        2212   4530
## Arkansas       2110   3378

Fancy Access: Conditions

Let’s look at the big states:

states[states$Population>10000,c('Area','Region')]
##                Area        Region
## California   156361          West
## Illinois      55748 North Central
## New York      47831     Northeast
## Ohio          40975 North Central
## Pennsylvania  44966     Northeast
## Texas        262134         South

Can also save this condition in a vector

bigstates = states$Population>10000
states[bigstates,'Abbr']
## [1] CA IL NY OH PA TX
## 50 Levels: AK AL AR AZ CA CO CT DE FL GA HI IA ID IL IN KS KY LA MA ... WY

Flow Control: For Loops

Generate the first 10 elements of the Fibonacci sequence

fibonacci = c(1,1)
for (i in 3:10){
  fibonacci[i] = fibonacci[i-1]+fibonacci[i-2]
}
fibonacci  
##  [1]  1  1  2  3  5  8 13 21 34 55

Flow Control: While Loops

A random walk

position = 0
total_steps = 0
while( abs(position) < 10){
  my.rand = runif(1)
  if (my.rand>0.5){
    step = 1
  } else {
    step = -1
  }
  position = position + step
  total_steps = total_steps + 1
}
total_steps
## [1] 26
position
## [1] 10

Bonus: Plotting

Basic Plotting: provide x and y data:

plot(states$Area,states$Population)

Bonus: Pretty Plots

We can pass additional parameters to plot that tell it what the plot should look like.

plot(states$Area,states$Population,xlab='Area',ylab='Population',main='Nice Plot!')

Bonus: More Points on One Plot

Let’s split the states up by population density

popdensity = states$Population/states$Area
dense  = (popdensity>median(popdensity))
sparse = (popdensity<=median(popdensity))

Then make a plot with both

plot(states$Area[dense],states$Population[dense],col='blue',xlab='Area',ylab='Population',main='Nice Colors!',xlim=c(0,6e5))
points(states$Area[sparse],states$Population[sparse],col='red',xlim=c(0,6e5))

Bonus: More Plots

You can produce side-by-side plots by setting a global parameter

par(mfrow=c(1,2))
plot(states$Area,states$Population,xlab='Area',ylab='Population',main='Super')
plot(states$Area,states$Income,xlab='Area',ylab='Income',main='Sweet!')