class 3: R basics

Prashant K

2024-01-18

Recap

  • class 2: Made our own histograms by binning data, understand mean, median and comparing 2 distributions

  • class 1: installed R, Rstudio and tidyverse package

    • If you already had these installed, check that they are the latest versions or re-install

      • R : version = 4.3.2 ; Rstudio (2022 or 2023 versions) ; tidyverse (just update it if you haven’t installed last week with install.packages('tidyverse')

Today’s class

  • Refresh: directory structure, .Rproject, R script

  • Introduce R datatypes: {tip: check datatype with class()}

    • Simple: numeric, character. Other simple: factor, logical

    • Compound: combination of simple datatypes: vector = c(x, y), list(x, y), data.frame() ; [won’t cover: matrix(), array()]

    • Good practices: Commenting code for documentation. # this code does x (if I get it to work)

  • Subset compound datatypes

  • Control statements: decision making: if() and repeat steps in a loop: for()

Key reference : base R cheatsheet.PDF

Complex plotting assignment

If you are well versed with R basics and are looking for some challenge, this slide is for you

  • Download the code we used to generate the data for the class2 activity

  • plot the individual data points from this code

  • Highlight the 30 points that your team got within this plot

    • With a different colour / outline / shape / anything fancy

  • This will take you ~2 lectures to finish so keep working on it

If you got any questions, you can ask me when I come around ; or the TAs if they are free from helping others learn R basics

Refresh / Rstudio overview

Refresh/setting up

  • .Rproject file: works from the current directory, saves history of R commands, and all scripts that were open last time.

  • .R script: A place where you save lines of R code and can run them all from top to bottom with one click or a command : source('..R')

Refresh / setting up

  • directory structure:

Directory structure

bash/terminal commands

  • mkdir: make directory

  • cd: change directory

  • ls: list files and directories

  • touch x.txt: make a empty files with any extension, ex: x.txt

Overview of all programming

  1. Take simple datatypes: 1 ; "apple", TRUE

  2. Combine into complex data types: c(1, 3, 5, 7) / c('apples', 'oranges') ; list( numbers, fruits..)

  3. Make decisions: if(x > 3) "apples are too sweet"

  4. Repeat actions with only slight changes: for(i in 1:5) do something with each i

  5. Make concise code by reusing parts as functions: do_magic <- function(x) {"x + magic here"}

Practice alongside me

Use the script: class3_R-basics.R from Canvas/files/ to follow along

Simple datatypes

Missing elements marked as NA or NaN

Making these datatypes in R ; assigning to a variable

  • numeric: x <- 35

  • character: y <- 'apples'

  • logical: a <- TRUE ; b <- F ; c <- T

  • factor:

Compound datatypes

combination of simple datatypes: vector = c(x, y), list(x, y), data.frame(x1 = x, x2 = y)

When printed, the list looks like this

$x
[1] 1 2 3 4 5

$y
[1] "a" "b"

A vector where entries are ordered and stored as “levels”

Use cases

  • Order of colours assigned when plotting stuff

  • Analyzing categorical questionnaires

  • Named vector: c('a' = 1, 'b' = 2)

  • List: list(x = 1:5, y = c('a', 'b'))

Recap: Lecture 3

  • Refresh: directory structure, .Rproject, R script, directory structure

  • Introduce R datatypes: {tip: check datatype with class()}

    • Simple: numeric, character. Other simple: factor, logical

    • Compound: combination of simple datatypes: vector = c(x, y), list(x, y), data.frame()

    • Good practices: Commenting code for documentation. # this code does x (if I get it to work)

  • Subset/indexing compound datatypes: x[3], x[x>0], dataframe[1,5]

  • Control statements: decision making: if() and repeat steps in a loop: for() or vectorize using functions

R basics continued

first 20 minutes of lecture 4 ; Date: 23/1/24

Let us quickly finish talking about

  • Vector subsetting / indexing

  • Programing : if..else ; for and function()

Note: Download the class3_R-basics.R file again from the website (I updated it.. You can rename last weeks file with the _old suffix … _old.R)

Subset compound datatypes

  • vectors: single element: x[3], multiple elements: x[x>0], by name x['a']?

  • Lists and dataframes:

    • whole row/column list$x1 or dataframe$x1 or dataframe[ , 2]

    • Single entry: dataframe[1,5]

Programming = control which statements are executed

  • Control statements: decision making: if() and repeat steps in a loop: for()

if..else :

if (condition){
  "Do something"
} else {
  "Do something different"
}

for() loops

for (variable in sequence){
  "Do something with the variable"
}