Useful commands in R

  • R is case sensitive
  • getwd() - Get current working directory
    • > getwd()
  • setwd() - Set current working directory
    • > setwd("C:/Vinay/Great Lakes/Great Lakes/Course/R/Working Directory")
  • Vector
    • Create a ordered vector of 100 elements from 1 till 100
      • > x <- 1: 100
    • Check the data type of the object
      • > class(5); # "numeric"
      • > class("Vinay"); # "character"
    • NOTE: 
      • Vector is a 1-dimensional array
      • Matrix is a 2-dimensional array
      • Arrays in R can be N-dimensional
      • Array index starts from 1
    • Function c for combine
      • Create a non-ordered vector with function c
        • > x <- c(3, 2, 1, 4, 5) 
        • > c(3, 2, 1 4, 5) -> x
        • > 1/x # 0.3334, 0.5, 1, .25, 0.2
        • > x[1] # 2
        • > c <- 2 * x #Vector arithmetic
        • > c # 6, 4, 2, 8, 10
  • Matrix
    • Function matrix
      • M <- matrix(1:9, 3, 3)
      • M # output below
      • M[1, 2] # 4
      • M[-1, -2] # output below. Matrix minus 1 row and 2 column
      • M <- matrix(c(1, 2, 3, 4), nrow = 2, ncol = 2)
    • Function rowSums, colSums
      • M <- matrix(c(1, 2, 3, 4), nrow = 2, ncol = 2)
      • M_rowSums <- rowSums(M); M_colSums <- colSums(M); 
      • M; M_rowSums; M_colSums; 
    • Adding a column using cbind i.e. column bind
      • M <- matrix(c(1, 2, 3, 4), nrow = 2, ncol = 2)
      • M; M <- cbind(M, c(5, 6)); M;
    • Adding a row using rbind i.e. row bind
      • M <- matrix(c(1, 2, 3, 4), nrow = 2, ncol = 2)
      • M; M <- rbind(M, c(5, 6)); M;
    • Matrix index
      • M <- matrix(c(1:25), nrow = 5, ncol = 5); M;
      • M[1, 2]; # 6 - element at 1st row and 2nd column 
      • M[1:3, 2:4]; # matrix subsection from 1st to 3rd rows and 2nd to 4th columns
      • M[,]; # similar to M
      • M[, 1]; # 1 2 3 4 5 - all row elements from 1st column
      • M[1, ]; # 1 6 11 16 21 - all column elements from 1 row
  • Data Frames
    • Two dimensional objects
    • Can hold Numeric, Character or Logical values
    • Within a column the data type is same for all elements
    • Different columns can have different data types
    • NOTE: Number of elements in each column should be same
    • Function data.frame
      • type <- c("Compact", "MiniVan"); mileage <- c(50, 40); price <- c(1000.00, 5050.00); no.cyl <- c(2, 3); 
      • cars <- data.frame(type, price, mileage, no.cyl)
      • cars; 
      • cars[1, 2] #1000
      • cars[1, 1:3] # Compact 1000 50
      • cars[, 1] # Compact MiniVan
      • cars[1, ] # Compact 1000 50 2
  • Lists
    • Can hold variety of objects including matrices, vectors, data frames, lists, etc
    • NOTE: Number of elements in each column need not be the same
    • Function list
      • vector <- c(1:10) # or vector <- 1:10
      • Matrix <- matrix(1:10, 5, 2)
      • myList <- list(vector, Matrix); myList;
      • myList[[1]][2]; # select 2 element from vector in list index 1 - Output = 2
      • myList[[2]][4]; # select row 4 and column 1 element from matrix in list index 2 - Output = 4
      • myList[[2]][1, 2]; # select row 1 and column 2 element from matrix in list index 2 - Output = 6
      • myList[[2]][1:3, 1:2]; # select subsection of matrix starting at row 1 till 3 and column 1 till 2 from matrix in list index 2 - Output as below:
  • Packages
    • Function install.packages
      • Use install.packages("PackageName") to install required packages
    • Function library
      • After installing the package, we can call the package in R code by library(PackageName)
  • Exercises
    • Create a vector containing elements 1 to 10 and 21 to 25
      • vector <- c(1:10, 21:25)
    • Create a matrix using the vector created above
      • Matrix <- matrix(vector, 5, 5)
    • Extract a 3 by 3 subset from the matrix created above
      • Subset <- Matrix[1:3, 1:3]; Subset;
  • Data Import
    • Function read.table/read.csv/read.xls
      • Use read.table("C:/Users/xyz/Desktop/myText.txt", header = TRUE, sep = ",") to read data from text(.txt) file
      • Use read.csv("C:/Users/xyz/Desktop/myCSV.csv", header = TRUE) to read data from Comma-Separated(.csv) file
      • Use read.xls("C:/Users/xyz/Desktop/myExcel.xls", header = TRUE) to read data from Excel(.xls) file. NOTE: Requires package gdata
    • Importing data from other Statistical software - Needs package foreign
      • read.spss - SPSS
      • read.dta - STATA
      • read.ssd - SAS
  • Working with data
    • Consider the sample data below saved in myCSV.csv file:
    • mydata <- read.csv("C:/Users/xyz/Desktop/myCSV.csv", header = TRUE)
    • Accessing columns
      • To read Price column data use mydata$Price
    • Function head 
      • head(mydata) #Gives by default, first 6 rows in the dataset 
      • head(mydata, 10) # First 10 rows in the dataset
    • Function tail
      • tail(mydata) #Gives by default, last 6 rows in the dataset
      • tail(mydata, 10) # Last 10 rows in the dataset
    • Function dim
      • dim(mydata) #Summary of the dataset. Output is number of rows and columns
    • Function summary
      • summary(mydata) #Summary of the dataset. Output contains Min, 1st Quadrant, Median, Mean, 3rd Quadrant and Max values for each column
    • Function view
      • view(mydata) #Another option to see the dataset
    • Function table
      • table(mydata$Fireplace) #Gives the frequency of value occurrence in a selected column Fireplace
        • Syntax: table(DatasetVariable$Column)/table(Column) if attach(DatasetVariable)
    • Subsetting/Filtering data
      • Use logical operators: ==, >, <, <=, >=, <> 
      • To select records with Fireplace set to 1, use mydata[mydata$Fireplace == 1, ]
        • Syntax: DatasetVariable[DatasetVariable$Column,] or DatasetVariable[Column,] if attach(DatasetName)
    • Function which 
      • mydata[which(mydata$Price <= 75000), ] #Select from the dataset only records with Price <= 75,000
      • mydata[which(mydata$Price <= 75000 & mydata$Bedrooms == 3), ] #Select from the dataset only records with Price <= 75,000 and 3 Bedrooms
      • Syntax: DatasetVariable[which(DatasetVariable$Columns separated with & filter), ]/DatasetVariable[which(Columns separated with & filter), ] if attach(DatasetVariable)
  • Exercises
    • Display only Price, Bedrooms and Age from the dataset
      • mydata[c(1, 4, 6)]
  • Help in R
    • ?
      • ?cbind
    • Function help
      • help(cbind)
  • Square using ^
    • x <- 1:10
    • xSquare <- x ^ 2 # x raise to the power 2. Output 1 4 9 16 25 36 49 64 81 100
  • Log using function log
    • x <- 1:10
    • xLog <- log(x) # Output 0.0000000 0.6931472 1.0986123 1.3862944 1.6094379 1.7917595 1.9459101 2.0794415 2.1972246 2.3025851
  • Function plot
    • Suppose we wish to see relationship between two variables:
      • x <- 1:10; y <- 2 * x - 8;
      • plot(x, y)
  • Function barplot
    • barplot(table(mydata$Bedrooms))
  • Function hist
    • Used to plot the frequencies of data
    • hist(mydata$Bedrooms) # Similar to barplot plotted above
    • hist(mydata$Bedrooms, main = "Bedrooms histogram") # Histogram with title
    • hist(mydata$Bedrooms, main = "Bedrooms histogram", xlab = "Bedrooms", ylab = "Frequency") # Histogram with title and X-Y axis label
    • hist(mydata$Bedrooms, main = "Bedrooms histogram", xlab = "Bedrooms", ylab = "Frequency", col = "lightBlue") # Histogram with title, X-Y axis label and color
  • Function boxplot
    • In Descriptive statistics, a box plot is a convenient way of graphically depicting groups of numerical data through their quartiles
    • boxplot(mydata$Price) # Create a default vertical boxplot
    • boxplot(mydata$Price, horizontal = TRUE) # Create a horizontal boxplot
    • boxplot(mydata$Price, horizontal = TRUE, main = "Price boxplot", xlab = "Price", col = "lightBlue") # Horizontal boxplot with title, X axis label and color
  • Package ggplot2
    • ggplot(mydata, aes(x = Price)) + geom_histogram() # with default 30 bins
    • ggplot(mydata, aes(x = Price)) + geom_histogram(stat = 'bin', binwidth = 10) # with specified bin size
  • Paneling Graphics using function par
    • We can partition the graphics panel
      • Syntax: par(mfrow = c(nrow, ncol)) # nrow = number of rows, ncol = number of columns
      • par(mfrow = c(2, 2));
      • x <- 1:10; y <- 2 * x - 8; plot(x, y); boxplot(mydata$Price, horizontal = TRUE); barplot(table(mydata$Bedrooms)); hist(mydata$Bedrooms, main = "Bedrooms histogram", xlab = "Bedrooms", ylab = "Frequency", col = "lightBlue");
  • Heat colors
    • hist(mydata$Bedrooms, main = "Bedrooms histogram", xlab = "Bedrooms", ylab = "Frequency", col = heat.colors(14));
  • Summary Stats
    • Function mean
      • mean(mydata$Price)
    • Function sd
      • sd(mydata$Price)
    • Function var
      • var(mydata$Price)
  • Function objects() and ls()
    • Lists all the objects created in a R session
    • objects(); ls();
  • Function rm()
    • Remove objects or functions
    • rm(cars); objects();
REFERENCES

Comments

Post a Comment