R tutorial summary
Wei Kuang, Deqing Fu, Yanbing Yi
2021/7/14
Introduction to R and R studio
The interface of R studio
The interface of R studio is divided into four parts. The top left block shows the code files. The bottom left block is console, where the code is excuted and the result is shown. You can see the files, the plots generated by code, the packages you’ve installed, and the help document in the bottom right block.
Opening/creating a file
If you would like to open an existing file, click the icon which looks like a folder.
If you would like to create a new file, click the icon at the left of the tool bar. You can choose the type of the file you would like to create. Next we will briefly introduce R script and R markdown.
R script
R script can only include code and comments. Comments starts following the # sign.
Clicking the run icon or press ctrl and enter at the same time can run the full code file. If you only want to run certain lines, select the lines and click run icon/press ctrl and enter simultaneously.
R markdown
R markdown is often used when people would like to write a report containing the results generated by R code. To generate a pdf document or html document, click the Knit icon.
The code blocks should always start and end as the following format:
1+1## [1] 2In the default setting, both the code block and its results will be shown in the final document. Code blocks have grey background. The results generated will follow right after the corresponding code block.
In the brace, you can set two parameters “echo” and “eval” to change what are included in the final document. If “echo” is set as FALSE, the code block will not appear while it will be excuted and the results will be shown. If “eval” is set as FALSE, the code block will be shown while it will not be excuted.
Packages
Installation
install.packages("glasso")R is an open source software, so there are many useful packages. If it is your fisrt time to use this package, you need to use the code above to install the package.
Import
library(glasso)Once the package is successfully installed, it will be there unless the package is uninstalled. However, the package needs to be imported each time when it is used.
Help document
Function
help(qnorm)
?(qnorm)The two lines above are equivalent for searching the usage of a certain function. In the help document, “usage” module shows how the function looks like. The arguments give out the definition of the input parameter of the function. “Value” tells what the function will return.
Package
help(base)You can use the same code to reach out to the help document of a package.
Data import and summary
Data import
In this section, we will introduce how to import the data in txt or excel document. The data set is the Female Horseshoe Crabs data from Example 1.5 in the book\(^{[1]}\) , which can be download from the website www.stat.ufl.edu/~aa/glm/data.
Txt file
Let us start with txt file.
crab_txt = read.table("D:/Stat@UChicago/2021Spring/consulting/Crabs.txt")
head(crab_txt) #only shows the first 6 rows of the data## V1 V2 V3 V4 V5 V6
## 1 crab y weight width color spine
## 2 1 8 3.050 28.3 2 3
## 3 2 0 1.550 22.5 3 3
## 4 3 9 2.300 26.0 1 1
## 5 4 0 2.100 24.8 3 3
## 6 5 4 2.600 26.0 3 3To import the data in txt file, we use the function “read.table” where the path of the file is put between the quotation marks. Please note that forward slash should be used in the path rather than the backslash. The data set is named as “crab_txt”, which means crab_txt will refer to this imported data set unless crab_txt is redefined.
You should note that the first row refers to the variables. However, the varaibles are supposed to be the names of columns instead of appearing as the first row. Thus, you can add a parameter “head” after the path of the file to make the modification.
crab_txt = read.table("D:/Stat@UChicago/2021Spring/consulting/Crabs.txt", head = TRUE)
head(crab_txt) #only shows the first 6 rows of the data## crab y weight width color spine
## 1 1 8 3.05 28.3 2 3
## 2 2 0 1.55 22.5 3 3
## 3 3 9 2.30 26.0 1 1
## 4 4 0 2.10 24.8 3 3
## 5 5 4 2.60 26.0 3 3
## 6 6 0 2.10 23.8 2 3Now each column is named by the corresponding variable. The first row becomes the data of the first female crab.
Excel file
crab_excel = read.csv("D:/Stat@UChicago/2021Spring/consulting/Crabs.csv", head = TRUE)
head(crab_excel)## crab y weight width color spine
## 1 1 8 3.05 NA 2 3
## 2 2 0 1.55 22.5 3 3
## 3 3 9 2.30 26.0 1 1
## 4 4 0 2.10 24.8 3 3
## 5 5 4 2.60 26.0 3 3
## 6 6 0 2.10 23.8 2 3It is similar to import a excel file using the function “read.csv”.
The dimension of data
dim(crab_excel)## [1] 173 6The function “dim” returns the dimension of the data set. In this example, the return value means this data set has 173 rows and 6 columns.
Summary of data
The value in the first row and fourth column in the excel file is deleted to better explain how to see missing values.
“summary” is commonly used to take a first look at the data.
summary(crab_excel)## crab y weight width color
## Min. : 1 Min. : 0.000 Min. :1.200 Min. :20.00 Min. :1.000
## 1st Qu.: 44 1st Qu.: 0.000 1st Qu.:2.000 1st Qu.:24.70 1st Qu.:2.000
## Median : 87 Median : 2.000 Median :2.300 Median :26.05 Median :2.000
## Mean : 87 Mean : 2.919 Mean :2.423 Mean :26.23 Mean :2.434
## 3rd Qu.:130 3rd Qu.: 5.000 3rd Qu.:2.800 3rd Qu.:27.62 3rd Qu.:3.000
## Max. :173 Max. :15.000 Max. :5.200 Max. :33.50 Max. :4.000
## NA's :1
## spine
## Min. :1.00
## 1st Qu.:2.00
## Median :3.00
## Mean :2.48
## 3rd Qu.:3.00
## Max. :3.00
## NA’s shows the number of the missing values in the corresponding column. For example, the first value in the fourth column is missing, so the NA’s for “width” column is 1.
Missing values
Furthermore, you might care about the location of the missing value in the “width” column. The following line can be used to identify which row the missing value locates.
which(is.na(crab_excel$width))## [1] 1Variable definition
There are basically two different variable assignment methods:
a = 2
b <- 3They are the same in the above setting.
However,
median(x = 1:10)## [1] 5.5x## Error in eval(expr, envir, enclos): object 'x' not foundIn the above case, x is defined with =, and the scope of the variable is local. Let’s see what happens when we switch to <-.
median(x <- 1:10)## [1] 5.5x## [1] 1 2 3 4 5 6 7 8 9 10When using <-, the variable x is globally defined. In most cases, these two assignment methods are the identical, but the R community seems to prefer <- to =.
Basic Operations
a <- 10; b <- 2;
a + b## [1] 12a - b## [1] 8a * b ## [1] 20a / b## [1] 5Data Types
Vector
The following are three different ways to define a vector.
vec1 <- c(2,3,2,4,1); vec1 #add the elements in order## [1] 2 3 2 4 1vec2 <- 2:6; vec2 #generate a sequence from 2 to 6 by step length 1## [1] 2 3 4 5 6vec3 <- seq(0,10, by=2); vec3 ##generate a sequence from 0 to 10 by step length 2## [1] 0 2 4 6 8 10Some useful functions
rev(vec1) #reverse a vector## [1] 1 4 2 3 2sort(vec1) #sort the elements of a vector## [1] 1 2 2 3 4unique(vec1) #delete the repeated elements in a vector## [1] 2 3 4 1Matrix
We introduce how to generate a matrix.
x = 1:16 #the elements in the matrix
#How to arrange the elements?
m1 = matrix(x, nrow = 4); m1 #the matrix has 4 rows, and the elements are arranged by column## [,1] [,2] [,3] [,4]
## [1,] 1 5 9 13
## [2,] 2 6 10 14
## [3,] 3 7 11 15
## [4,] 4 8 12 16m2 = matrix(x, nrow = 2, ncol = 8); m2 #the matrix has 2 rows and 8 columns, and the elements are arranged by column## [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8]
## [1,] 1 3 5 7 9 11 13 15
## [2,] 2 4 6 8 10 12 14 16Extract elements from the matrix
m1[1,] #the first row## [1] 1 5 9 13m1[,3] #the third column## [1] 9 10 11 12m1[2,2] #the entry at the second row and second column## [1] 6Basic matrix operations
t(m1) # transpose## [,1] [,2] [,3] [,4]
## [1,] 1 2 3 4
## [2,] 5 6 7 8
## [3,] 9 10 11 12
## [4,] 13 14 15 16det(m1) # determinant## [1] 0diag(m1) # diagonal elements## [1] 1 6 11 16diag(4) # identity matrix of 4 x 4## [,1] [,2] [,3] [,4]
## [1,] 1 0 0 0
## [2,] 0 1 0 0
## [3,] 0 0 1 0
## [4,] 0 0 0 1String
s = "hello world!"; s## [1] "hello world!"# string concatenation
s = paste(s, "I'm new to R"); s## [1] "hello world! I'm new to R"# string formatting
hour = 9; minute = 30;
s = sprintf("Hello! It's %d:%d a.m. now", hour, minute); s## [1] "Hello! It's 9:30 a.m. now"Data Frame
# generate a data frame
survey <- data.frame("index" = c(1, 2, 3, 4, 5),
"age" = c(24, 25, 42, 56, 22))
survey## index age
## 1 1 24
## 2 2 25
## 3 3 42
## 4 4 56
## 5 5 22# extract a column
survey$index## [1] 1 2 3 4 5survey$age ## [1] 24 25 42 56 22## adding another column
survey$edu <- as.factor(c("college", "hs", "hs", "doctorate", "college"))
survey ## index age edu
## 1 1 24 college
## 2 2 25 hs
## 3 3 42 hs
## 4 4 56 doctorate
## 5 5 22 college## summary of data frame
summary(survey)## index age edu
## Min. :1 Min. :22.0 college :2
## 1st Qu.:2 1st Qu.:24.0 doctorate:1
## Median :3 Median :25.0 hs :2
## Mean :3 Mean :33.8
## 3rd Qu.:4 3rd Qu.:42.0
## Max. :5 Max. :56.0Plots
In this section, we will show how to generate some plots using R code. The built-in data set iris will be used as an example.
Let us first take a look at the iris data set.
data <- irishead(data)## Sepal.Length Sepal.Width Petal.Length Petal.Width Species
## 1 5.1 3.5 1.4 0.2 setosa
## 2 4.9 3.0 1.4 0.2 setosa
## 3 4.7 3.2 1.3 0.2 setosa
## 4 4.6 3.1 1.5 0.2 setosa
## 5 5.0 3.6 1.4 0.2 setosa
## 6 5.4 3.9 1.7 0.4 setosasummary(data)## Sepal.Length Sepal.Width Petal.Length Petal.Width
## Min. :4.300 Min. :2.000 Min. :1.000 Min. :0.100
## 1st Qu.:5.100 1st Qu.:2.800 1st Qu.:1.600 1st Qu.:0.300
## Median :5.800 Median :3.000 Median :4.350 Median :1.300
## Mean :5.843 Mean :3.057 Mean :3.758 Mean :1.199
## 3rd Qu.:6.400 3rd Qu.:3.300 3rd Qu.:5.100 3rd Qu.:1.800
## Max. :7.900 Max. :4.400 Max. :6.900 Max. :2.500
## Species
## setosa :50
## versicolor:50
## virginica :50
##
##
## Types of Plots
- Histograms
- Barplots
- Boxplots
- Scatterplots
Histograms
The most basic usage of “hist” function is inputting the data for the histogram. In this setting, the number of the bins will be automatically decided by the function.
hist(data$Sepal.Length)
There are many parameters to adjust the histogram. Please see the following code block for details.
hist(data$Sepal.Length, #the data for histogram
main = "Histogram of Sepal Length", #title of the histogram
xlab = "Sepal Length", #the name of x axis
breaks = 20, #the number of the bins in histogram
col = "grey90", #color of the fillings in the bins
border = "darkblue")#color of the border of the bins
* breaks is specific to hist() * By default R will attempt to intelligently guess a good number of breaks * Entering an integer will give a suggestion to R for how many bins to use for the histogram
Barplots
Barplots are useful to visualize the summary of a categorical variable
table(data$Species)##
## setosa versicolor virginica
## 50 50 50“table” is a function which returns the counts. In the code block above, it shows the numbers of flowers for three species.
The most basic usage of “barplot” is entering the counts of all the classes.
barplot(table(data$Species))
There are also parameters to refine the barplots. Please see the following code block for details.
barplot(table(data$Species), #the counts of all classes
names = c("setosa", "versicolor", "virginica"), #the names of all bars
xlab = "Species", #the name of x axis
main = "Barplot of Species", #the title of the barplot
ylim = c(0,60), #the range of y axis
col = c("darkblue", "darkred", "darkgreen"), #the colors of all bars
border = "gold")#the color of the borders of the bars
If you woule like to rotate the bar plot, you can set the parameter “horiz”.
barplot(table(data$Species), #the counts of all classes
names = c("setosa", "versicolor", "virginica"), #the names of all bars
ylab = "Species", #the name of y axis
main = "Barplot of Species", #the title of the barplot
xlim = c(0,60), #the range of x axis
horiz = T,
col = c("darkblue", "darkred", "darkgreen"), #the colors of all bars
border = "gold")#the color of the borders of the bars
Boxplots
We can use boxplots to display the relationship of a categorical variable and a numeric variable.
Boxplot can be used to show the distribution of a numeric data set in some sense.
boxplot(data$Sepal.Length)
The boxplot can also reflect the data of different classes simultaneously.
boxplot(Sepal.Length ~ Species, data = data)
The x axis is classified by the species, and boxplot of the Sepal.Length of each sepicies are laid in order. Inputting more parameters can refine the boxplot. The following is an example:
boxplot(Sepal.Length ~ Species, data = data,
xlab = "Species", #the name of x axis
ylab = "Sepal Length", #the name of y axis
main = "Sepal Lengh vs Species", #the title of the boxplot
col = "grey90", #the color of the fillings in the box
border = "darkblue") #the color of the border of the boxplot
Scatterplots
If only the data of one variable is input, the x axis will be the index of each data point automatically.
plot(data$Sepal.Length)
Scatterplot can also reflect the relationship of two variables. In the following example, the x and y coordinates of each point represent its sepal width and sepal length respectively.
plot(Sepal.Length ~ Sepal.Width, data = data)
plot(data$Sepal.Width, data$Sepal.Length)
#The two ways above are equivalent.plot(Sepal.Length ~ Sepal.Width, data = data,
xlab = "Sepal Width", #the name of x axis
ylab = "Sepal Length", #the name of y axis
main = "Sepal Lengh vs Sepal Width", #the title of the plot
col = "darkblue") #the color of the points
Line plots
In the “plot” function, setting the parameter “type” as “l” helps return a line plot.
plot(data$Sepal.Length, #the data
type = "l", #the type of the plot is line plot
col = "darkblue", #the color of the line
lwd = 1, #the width of the line
lty = 1, #the type of the line
ylab = "Sepal Length", #the name of y axis
main = "Sepal Length") #the title of the line plot
Different lines can be added on one plot.
plot(data$Sepal.Length, #the data
type = "l", #the type of the plot is line plot
col = "darkblue", #the color of the line
lwd = 1, #the width of the line
lty = 1, #the type of the line
ylim = c(0, 8),
ylab = "y", #the name of y axis
main = "Sepal length and width") #the title of the line plot
lines(data$Sepal.Width,
col = "darkred",
lwd = 1,
lty = 2)
Legend is often used to distinguish different lines.
plot(data$Sepal.Length, #the data
type = "l", #the type of the plot is line plot
col = "darkblue", #the color of the line
lwd = 1, #the width of the line
lty = 1, #the type of the line
ylim = c(0, 8),
ylab = "y", #the name of y axis
main = "Sepal length and width") #the title of the line plot
lines(data$Sepal.Width,
col = "darkred",
lwd = 1,
lty = 2)
legend("topleft", #the location of the legend
legend = c("sepal length", "sepal width"), #the variables corresponding to all lines
col = c("darkblue", "darkred"), #the color of the lines
lty = c(1,2), #the type of the lines
cex = 0.5) #the size of the legend box
Reference
[1] Alan Agresti. Foundations of linear and generalized linear models. John Wiley & Sons, 2015.