Intro to Practical Computing in R

Introduction to R

Sept 15

Version control review

Read page two of this [git cheatsheet] if you haven’t already! Read it every so often to remind yourself of what the point of version control is and how to use it most effectively.

The take homes from lecture:

  • commit often
  • commit related changes
  • commit messages are for you
  • version control is not a replacement for backups

Some basics of R and R syntax

This is a follow-up to the work you have done using the Swirl lessons 1 and 2. We will beging to familiarise ourselves we will learn about basic R syntax, some R functionality, and learn how to interpret the output of str().

Today, we will familiarise ourselves with RStudio’s layout, meet our dataset, and meet the function str().

We’re going to run through some things relatively quickly.

Basic mathematical operations *modified from swc

You can use R just like a calculator. 

2 + 2

Order of operations are followed. 

(3 + 4 * (10 - 5)^2)

Scientific notation

Scientific notation can and is used in R. These three values are equivalent. 

0.0003
3 * 10^(-4)
3e-4

Functions

R includes mathematical functions. Note in the examples below that log() is the natural logarithm in R, if you would like to use a logarithm with a different base, such as base 10, you need to include that explicitly, as demonstrated. 

log(2)  # natural log
log10(2)  # base-10 log

exp(0.5)  # e^(1/2)

Comparisons 

2 == 2

2 == 3

2 != 3
2 != 2

2 < 3
2 > 3

2 < 2

2 <= 2
2 <= 3

Floating point numbers

We can see that we can check equivalence using the == operator. However, you should be aware of an interesting behaviour. I’ll expand the explanation soon and for those who want to know more, check it out on google. But essentially storing decimal values with perfect precision can cause difficulties on a computer. This can make doing comparisons tricky, where numbers that look and really are equal, do not satisfy the equivalence ==. To really test whether values are equal (this does come up sometimes), you are best off to use the all.equal() function in cases where you suspect you may run across floating point issues.

Try this out

0.0003 == 3 * 10^(4)

Now try this

all.equal(0.0003, 3 * 10^(-4))

Tip: These are considered operators. They include things like ‘or’, ‘and’, ‘divide’, ‘less than’. This is useful to know when googling!

Tip from swc gapminder: Instead you should use the all.equal function. (make sure to bring up later in the course)

Exercise 1

  1. If you see this in your console, what happened? And how can you fix it? 
> 1 + 
+
  1. What happens when you try and run the following? What are some solutions? 
e^(1/4)

##### Answers

  1. You can use the Esc key (when working in RStudio). If you are in a terminal you can use ctrl + c (this also works for killing any process in a terminal - e.g., you forgot to close parenthases when writing your git commit message…).

  2. The e is not found! This is because e is not a function. To solve this problem, you could assign e to equal e: e <- 2.718 OR you could use the built-in function exp(1/4).

Assigning variables

You will also have seen this in swirl. Variables are objects. As you’ve seen in Swirl, they can be single values, a list of values, or even contain nested objects. Like a list within a list. When naming variables, you typically want to keep the names short (since you have to type them out) but descriptive. I prefer to use use snake case or camel case rather than separating words by dots.

So you’ve now seen that you can assign a variable using the <- operator. Has anyone found the keyboard shortcut to do this in RStudio? (option + -).

Sometimes it is okay to just use very short variable names like x, but in cases where you have variables that really mean something, it can be more informative to you, or someone who will be reading your code later, to give it a name that is self explanatory. 

x <- 16 


# If we were to have a list of temperature values (obviously this is in celsius!), 'temp' might be an appropriate name
temp <- c(16, 12, 20, 13, 10)

Basics of syntax - functions

You will use functions in almost everything you do in R. Later in the semester we’re going to learn how to write them (which really isn’t that bad). But for now, you need to be familiar with what a function is. The functions that you will use will have the following format.

function(arg1, arg2, arg3, … )

A function can take arguments. Some functions do, some don’t. You can refer to the function’s help file to find out what arguments are required.

I use the function args(foo) (where foo is your function of interest) to get a quick look into what arguments a function takes. While super handy, as we saw in class, it doesn’t always give you all the information that you want, so in that case you’ll have to look into the R help files.

Helping yourself

Today we looked at the help file for the mean function mean(). Coming soon, screenshots and description of a help file

There are multiple ways to access the helpfiles. Can you find them? One way we can do it is through the console using ?. 

?mean()

The others are through the RStudio help interface/search, or if you just google for the functions.