Book Image

Hands-On Exploratory Data Analysis with R

By : Radhika Datar, Harish Garg
Book Image

Hands-On Exploratory Data Analysis with R

By: Radhika Datar, Harish Garg

Overview of this book

Hands-On Exploratory Data Analysis with R will help you build a strong foundation in data analysis and get well-versed with elementary ways to analyze data. You will learn how to understand your data and summarize its characteristics. You'll also study the structure of your data, and you'll explore graphical and numerical techniques using the R language. This book covers the entire exploratory data analysis (EDA) process—data collection, generating statistics, distribution, and invalidating the hypothesis. As you progress through the book, you will set up a data analysis environment with tools such as ggplot2, knitr, and R Markdown, using DOE Scatter Plot and SML2010 for multifactor, optimization, and regression data problems. By the end of this book, you will be able to successfully carry out a preliminary investigation on any dataset, uncover hidden insights, and present your results in a business context.
Table of Contents (17 chapters)
Free Chapter
1
Section 1: Setting Up Data Analysis Environment
7
Section 2: Univariate, Time Series, and Multivariate Data
11
Section 3: Multifactor, Optimization, and Regression Data Problems
14
Section 4: Conclusions

Exploring graphically the dataset

In this section, we will focus on exploring the dataset graphically using a DOE scatter plot, a DOE mean plot, a DOE standard deviation plot, and a contour plot. Let's focus on each of them in turn:

  1. In this step, we will depict the scatter plot in two ways. A scatter plot shows the relationship between wt and mpg as follows:
> plot(Autompg$weight , Autompg$mpg, xlab = 'Weight of Cars', ylab = 'Miles per Gallon', main = 'Scatter Plot for MTCars Weight Vs MPG')

This gives us the following output plot:

The alternative way to depict the scatter plot is with the help of the ggplot2 package or library, which is achieved by executing the following command:

> library(ggplot2)
> ggplot(data=Autompg,aes(x=weight, y=mpg)) + geom_point() + theme_minimal()

This gives us the following output plot:

  1. This step...