Cluster Analysis : Motivation, Application, Theory & SAS procedure

Cluster analysis is one of the most popular techniques used in data mining for marketing needs. The idea behind cluster analysis is to find natural groups within data in such a way that each element in the group is as similar to each other as possible. At the same time, the groups are as dissimilar to other groups as possible.

This course will make you adept in the concepts of cluster analysis and you will also learn how to use SAS (or similar other statistical software) to conduct hierarchical and non-hierarchical clustering and how to interpret the output.

Who should enroll for this course?

• This course is for statistics or analytics professionals / students who need to apply this in their work but get confused with jargon and are overwhelmed with a plethora of materials available on the topic.

Why should one enroll in this particular course?

• You will understand all the concepts in crystal clear manner
• The course is a good mix of theory and practical side of cluster analysis
• You will learn to the extent that you will be able to apply your learning directly after the course
• You will learn the SAS commands for cluster analysis and detail explanation of it's output.

Prerequisite:

• Basic understanding of SAS platform

Course Package:

• 26 Video Presentations (powerpoint presentations with voice)
• 04 PDFs’
• 03 Excel Workbook (containing SAS codes, SAS output and detail explanation)

Key Takeaways:

• Crystal clear understanding of cluster analysis and various associated terms like link function, distance function, k-means algorithm, dendogram, scree plot etc.
• Learn how to use hierarchical and non hierarchical method; What is the SAS command / procedure for the same; How to interpret the SAS output

Course Content:

• Motivation –
  1. Where one applies cluster analysis. Why one should learn cluster analysis?
  2. How it is different from objective segmentation (CHAID / CART)
• Statistical foundation and practical application: Understand
  1. Cluster Analysis – high level view
  2. Different type of cluster analysis
  3. Hierarchical clustering –
  • Agglomerative or Divisive technique
  • Dendogram – What it is? What does it show?
  • Scree plot - How to decide about number of clusters
  • How to use SAS command to run hierarchical clustering
  • When and why does on need to standardize the data?
  • How to understand and interpret the output
• Non-hierarchical clustering (K means clustering).
  • Why do we need k means approach
  • How does it work?
  • How does it iterate?
  • How does it decide about combining old clusters?
  • How to use SAS command to run hierarchical clustering
  • When and why does on need to standardize the data?
  • How to understand and interpret the output