Book Image

Practical Predictive Analytics

By : Ralph Winters
Book Image

Practical Predictive Analytics

By: Ralph Winters

Overview of this book

This is the go-to book for anyone interested in the steps needed to develop predictive analytics solutions with examples from the world of marketing, healthcare, and retail. We'll get started with a brief history of predictive analytics and learn about different roles and functions people play within a predictive analytics project. Then, we will learn about various ways of installing R along with their pros and cons, combined with a step-by-step installation of RStudio, and a description of the best practices for organizing your projects. On completing the installation, we will begin to acquire the skills necessary to input, clean, and prepare your data for modeling. We will learn the six specific steps needed to implement and successfully deploy a predictive model starting from asking the right questions through model development and ending with deploying your predictive model into production. We will learn why collaboration is important and how agile iterative modeling cycles can increase your chances of developing and deploying the best successful model. We will continue your journey in the cloud by extending your skill set by learning about Databricks and SparkR, which allow you to develop predictive models on vast gigabytes of data.
Table of Contents (19 chapters)
Title Page
About the Author
About the Reviewers
Customer Feedback

Variable reduction/variable importance

Variable reduction techniques allow you to reduce the number of variables that you need to specify to a model. We will discuss three different methods to accomplish this.

  1. Principal Components Analysis (PCA).
  2. All subsets Regression.
  3. Variable Importance.

Principal Components Analysis (PCA)

Principle Components Analysis (PCA) is a variable reduction technique, and can also be used to identify variable importance. An interesting benefit of PCA is that all of the resulting new component variables will all be uncorrelated with each other. Uncorrelated variables are desirable in a predictive model since too many correlated variables confound predictions and make it difficult to tell which of the independent variables have the most influence. So, if you first perform an exploratory analysis of your data and you find that a high number of correlations exist, this would be a good opportunity to apply PCA.


Models can tolerate some degree of correlated variables...