Begginer guide: get familar with coding in R

rm(list = ls())

graphics.off()

age = c(45,75,65,85,45,32,57,65,52,45,65,32,89,45,66)

R operators

R provides standard arithmetic operators for addition , substraction, multiplication, division, and exponential. The basic symbols used by R to perform mathematical operations are called operators. Some of the mathematical operators are listed in table 1. Note that the forward slash (/) is used for division because it’s similar to the division line that you would use when writing a fraction.

require(magrittr)

data.frame(Symbol = c("+", "-", "*", "/"), 
           Operation = c("Addition", "Substraction","Multiplication","Division")) %>%
  knitr::kable(format = "html", align = "l", caption = "Basic R operators")

Table 1: Basic R operators

Symbol	Operation
	Addition
	Substraction
	Multiplication
/	Division

Because of the convinient, we will use Rstudio console, just write an expression 2 + 3 and click Enter

2+3

[1] 5

As we expected, R returns the answer as 5. Unlike other programming languages, coding in R does not need to terminate the expression or lines with a semicolon.

10-4

[1] 6

23*2

[1] 46

8/2

[1] 4

5^2

[1] 25

5%%3

[1] 2

5%%5

[1] 0

Precedence

R can be used to express complicated mathematical formulas. For anyone unfamiliar with writing formulas on computers, it is important to recognize that R will make assumptions about which part of the formulas to compute first. This order of operations is referred as predecence. Multiplication and division have a higher order than addition and subtraction. For example guess what is the answer for the mathematical expression in the chunk below.

5 - 3 + 8 / 2 * 3

The answer should be 14. According precedence, the number 8 is first divided by 2 to get 4 as product which is multiplied by 3 to obtain 12 and then 5 is added to get 17, which is substracted by 3 to get 14 as final result. We use parentheses in a R to control the order of operations like in the expression below. Can you make a mental calculation and provide the answer before you click enter for the expression in the chunk below?

5 - (3 + 8) / 2 * 3

The result of the expression is -11.5 and not 14 because the parentheses tell R to do the operation in the parentheses firs and then do the operation outside the parentheses.

Variables

We can create expression using variables. for instance, we assign the value 5 to the variable x and evaluate the square of x using the exponential ^ operator.

x = 5
x^2

[1] 25

R has many types of variables that store different kinds of data in different ways. Example you can store a list of numbers or text

days= c("Monday", "Tuesday", "Wednesday")
speed  = c(128, 158, 89)

R has peculiar syntax when it comes to variable names. The dot character . has a completely different meaning as compared to other programming languages. In R, we can use . in the variable names, so x.1 and x_1 are perfectly valid. In practice, the dot operator is used as a visual separator in variable names, similar to underscore in most other programming languages.

Naming conventions

• Name must begin with a letter. Do not use number, dollar sign or underscore
• R is case sensitive; average and Average are two different objects in R
• Use descriptive names
• if the name is made of more than one word, used period (. or underscore to separate the words)

Functions

Functions in R are first class objects, which means that they can be treated much like any other R object. Importantly,

• Functions can be passed as arguments to other functions
• Functions can be nested, so that you can define a function inside of another function

The return value of a function is the last expression in the function body to be evaluated.

R functions arguments can be matched positionally or by name. So the following calls to sd are all equivalent

data = rnorm(25)
sum(data)

[1] 2.875542

sd(data)

[1] 0.9071268

mean(data)

[1] 0.1150217

median(data)

[1] 0.1755854

Essential functions

R has many built-in functions that can be used fo a great variety of tasks. These can be suplemented with packages, which contains more functions bundled in one document. Here is a list of common and widely used functions + rep() — repeates a value some number of times to make a list + seq() — creates a sequence of values between a start and end number and spaced at certain interval + aggregate() — used to bin data by condition + table() — used to summarise categorical data + plot() — graphical plots of data + hist() — a function for plotting a histogram + boxplot() — a function for plotting boxplot + mean() — compute arithmetric mean + sd() — compute arithmetric standard deviation + sum() — compute the total of the set of elements + length() — function to count the number of elements in a vector

Installing packages

R comes with a standard set of packages. Others are available for download and installation. The primary of stable package is the CRAN. In R you can install a package from CRAN with an install.packages("packagename") function that allows you to install the package you want to use into R. In have already installed the ggplot2 packages in my machine, so if you want to install in your machine you can simply uncomment the chunk below by removing the hash tag (#)

## install.packages("ggplot2")
## install.packages("dplyr")
## install.packages("lubridate")
## install.packages("factoextra")
## install.packages("readxl")
## install.packages("kableextra")
## install.packages("haven")
## install.packages("readr")

Loading packages

Once package is downloaded and installed in your computer, you have to them into the session to access its functions and resources of the package. Yu can load the packages you want ot use with ether library() or required() function.

require(dplyr)
library(readr)
require(lubridate)
library(readxl)
require(haven)
library(ggplot2)
require(kableExtra)

rm(list = ls())
graphics.off()

Data Types

R is a flexible language that allows to work with different kind of data format [@bradley]. This inluced integer, numeric, character, complex, dates and logical. The default data type or class in R is double precision—numeric. In a nutshell, R treats all kind of data into five categories but we deal with only four in this book. Before proceeding, we need to clear the workspace by typing rm(list = ls()) after the prompt in the in a console.

1. Integers:Integer values do not have decimal places. They are commonly used for counting or indexing.

aa = c(20,68,78,50)

You can check if the data is integer with is.integer() and can convert numeric value to an integer with as.integer()

is.integer(aa)

[1] FALSE

You can query the class of the object with the class() to know the class of the object

class(aa)

[1] "numeric"

Although the object bb is integer as confirmed with as.integer() function, the class() ouput the answer as numeric. This is because the defaul type of number in r is numeric. However, you can use the function as.integer() to convert numeric value to integer

class(as.integer(aa))

[1] "integer"

2. Numeric: The numeric class holds the set of real numbers — decimal place numbers. The numeric class is more general than the integer class, and inclused the integer numbers.

These could be any number (whole or decimal number). You can check if the data is integer with is.integer()

bb = c(12.5, 45.68, 2.65)
class(bb)

[1] "numeric"

is.numeric(bb)

[1] TRUE

3. Strings: These collection of characters. This often are text data like names. You can check if the data is integer with is.character()

kata = c("Dege", "Mchikichini", "Mwembe Mdogo", "Cheka")
class(kata)

[1] "character"

4. Factor: These are strings from finite set of values. For example, we might wish to store a variable that records gender of people. You can check if the data is factor with is.factor() and use as.factor() to convert string to factor

sex = c("Male", "Female", "Male", "Male", "Female")
sex = as.factor(sex)
class(sex)

[1] "factor"

levels(sex)

[1] "Female" "Male"  

Often times we need to know the possible groups that are in the factor data. This can be achieved with the levels() function

levels(sex)

[1] "Female" "Male"  

levels(kata)

NULL

Often we wish to take a continuous numerical vector and transform it into a factor. The function cut() takes a vector of numerical data and creates a factor based on your give cut-points. Let us make a fictional income of 508 people with rnorm() function.

income = rnorm(n = 508, mean = 500, sd = 80)
hist(income, col = "green", main = "", las = 1, xlab = "Individual Income")

#mosaic::plotDist(dist = "norm", mean = 500, sd = 80)

We can now breaks the distribution into groups and make a simple plot as shown in figure 2, where those with income less than 400 were about 50, followed with a group with income range between 400 and 500 of about 200 and 250 people receive income above 500

group = cut(income, breaks = c(300,400,500,800),
            labels = c("Below 400", "400-500", "Above 500"))
is.factor(group)

[1] TRUE

levels(group)

[1] "Below 400" "400-500"   "Above 500"

barplot(table(group), las = 1, horiz = FALSE, col = c("blue", "red", "blue"), ylab = "Frequency", xlab = "Group of Income")

data = data.frame(group, income)

5. Logicals: This is a special case of a factor that can only take on the values TRUE and FALSE. R is case-sensitive, therefore you must always capitalize TRUE and FALSE in function in R.

6. Date and time

Vectors

Ofen times we want to store a set of numbers in once place. One way to do this is using the vectors in R. Vectors store severl numbers– a set of numbers in one container. let us look on the example below

id = c(1,2,3,4,5)
people = c(158,659,782,659,759)
street = c("Dege", "Mchikichini", "Mwembe Mdogo", "Mwongozo",  "Cheka")

Notice that the c() function, which is short for concatenate wraps the list of numbers. The c() function combines all numbers together into one container. Notice also that all the individual numbers are separated with a comma. The comma is reffered to an an item-delimiter. It allows R to hold each of the numbers separately. This is vital as without the item-delimiter, R will treat a vector as one big, unsperated number.

Indexing the element

One advantage of vector is that you can extract individual element in the vector object by indexing, which is accomplished using the square bracket as illustrated below.

id[5]

[1] 5

people[5]

[1] 759

street[5]

[1] "Cheka"

Apart from extracting single element, indexing allows to extract a range of element in a vector. This is extremely important because it allows to subset a portion of data in a vector. A colon operator is used to extract a range of data

street[2:4]

[1] "Mchikichini"  "Mwembe Mdogo" "Mwongozo"    

Adding and Replacing an element in a vector

It is possible to add element of an axisting vecor. Here ia an example

id[6] = 6
people[6] = 578
street[6] = "Mwongozo"

Sometimes you may need to replace an element from a vector, this can be achieved with indexing

people[1] = 750

Number of elements in a vector

Sometimes you may have a long vector and want to know the numbers of elements in the object. R has length() function that allows you to query the vector and print the answer

length(people)

[1] 6

data.tb = tibble::tibble(time, distance)
tibble::is_tibble(data.tb);class(data.tb) 

[1] TRUE

[1] "tbl_df"     "tbl"        "data.frame"

## create a data.frame
data.df = data.frame(time, distance)
## create a tibble
data.tbl = tibble::as_tibble(data.df)
## printout 
tibble::is_tibble(data.df); tibble::is_tibble(data.tb)

[1] FALSE

[1] TRUE

Column Names

Unlike data.frame that is strict on column names, tibble need not be valid R variable names. They can contain unusual characters like a space or a smiley but be enclosed in ticks.

tibble::tibble(id = 1:10,
               ":" = rep(c("NE", "SE"), each = 5), 
               "-" = rnorm(10, 25,2), 
               chl = rnorm(10,.25,.01))

# A tibble: 10 x 4
      id `:`     `-`   chl
   <int> <chr> <dbl> <dbl>
 1     1 NE     22.4 0.246
 2     2 NE     24.4 0.250
 3     3 NE     22.2 0.258
 4     4 NE     24.8 0.263
 5     5 NE     22.1 0.239
 6     6 SE     28.4 0.247
 7     7 SE     21.2 0.253
 8     8 SE     26.4 0.246
 9     9 SE     26.5 0.255
10    10 SE     27.4 0.242

However, when we use data.frame()instead of tibble() with the same arguments, we notice that the data.frame() function has modified the column names.

data.frame(id = 1:10,
           ":" = rep(c("NE", "SE"), each = 5), 
           "-" = rnorm(10, 25,2), 
           chl = rnorm(10,.25,.01))

   id X.     X..1       chl
1   1 NE 23.54136 0.2392456
2   2 NE 26.30564 0.2516395
3   3 NE 28.08723 0.2420568
4   4 NE 26.36949 0.2482949
5   5 NE 23.30528 0.2553076
6   6 SE 27.58529 0.2332887
7   7 SE 25.80831 0.2552407
8   8 SE 22.65104 0.2502862
9   9 SE 25.10100 0.2497607
10 10 SE 24.84666 0.2503427

Add rows and columns

tibble allows you to add column and rows. Let us add the unique ID in the data.tb we created earlier using add_column() function. This function allows us to specify wher we want to place the column we will create. for this case we want the id variable to be the first, hence we specify .before = 1 argument.

tibble::add_column(.data = data.tb, id = 1:10, .before = 1)

# A tibble: 10 x 3
      id time                distance
   <int> <dttm>                 <dbl>
 1     1 2018-11-20 06:35:25       20
 2     2 2018-11-20 06:52:05       85
 3     3 2018-11-20 07:08:45       45
 4     4 2018-11-20 07:25:25       69
 5     5 2018-11-20 07:42:05       42
 6     6 2018-11-20 07:58:45       52
 7     7 2018-11-20 08:15:25        6
 8     8 2018-11-20 08:32:05       45
 9     9 2018-11-20 08:48:45       36
10    10 2018-11-20 09:05:25        7

We can also add row(s) in tibble with the add_row() function and specify at which row location we want to insert the rows. For instance, in the code below, we command the rows to be inserted from row 5

tibble::add_row(.data = data.tb, 
                time = seq(lubridate::dmy(120514),lubridate::dmy(150514), by = "day"), 
                distance = rnorm(4, 23,6), 
                .before = 5)

# A tibble: 14 x 2
   time                distance
   <dttm>                 <dbl>
 1 2018-11-20 06:35:25     20  
 2 2018-11-20 06:52:05     85  
 3 2018-11-20 07:08:45     45  
 4 2018-11-20 07:25:25     69  
 5 2014-05-12 00:00:00     22.5
 6 2014-05-13 00:00:00     18.4
 7 2014-05-14 00:00:00     15.1
 8 2014-05-15 00:00:00     16.3
 9 2018-11-20 07:42:05     42  
10 2018-11-20 07:58:45     52  
11 2018-11-20 08:15:25      6  
12 2018-11-20 08:32:05     45  
13 2018-11-20 08:48:45     36  
14 2018-11-20 09:05:25      7