×

Warning

The RokSprocket Module needs the RokSprocket Component enabled.

Principal Component Analysis

Principal Component Analysis

Learn about this data reduction technique to identify, quantify and visualise the correlation in set of measurements. PCA provides insightful data visualisation tools. Learn about innovative applications.

Course Details

Learning Objectives

Learning Objectives:

Upon completion of this module, participants will know:

    • Why and when to use PCA
    • The required data layout for running PCA in software packages
    • How to choose the options based on data features
    • How to interpret the software output
    • How to produce informative 2D data summaries
    • How to map data, interpret results and draw conclusions

 

Target Audience

Target Audience:

This module is intended for all scientific staff who collect large datasets and who wish to graphically summarizing them and identify redundancy for the purpose of data reduction.

Prerequisite

Prerequisite:

This training course introduces the important ideas in statistics and multivariate data analysis. It assumes that participants have no previous knowledge of statistics or that they have not used it for a long time.

Course Outline

Course Outline:
    • Key Statistical Concepts for PCA
    • Traditional Methods for Analyzing a Set of Measurements
    • Fundamentals of PCA
    • Determining Principal Components (PCs)
    • Software Output Through a Case Study
    • Summary Statistics and Other Table Output
      • Variability Explained by Each PC
      • Correlation & Variance/Covariance Matrix
      • Loadings of the PCs
      • Coordinates on the PCs
    • Graphical Output
      • The Scree Plot
      • Components Loadings
      • Correlation Circle
      • Object Map on the New PCs
      • Biplot
    • Specific Analysis Issues
      • Choosing an Adequate Measure of Variability
      • Selecting a Subset of Components
    • A Step-by-Step Approach to PCA
    • Case Studies
    • Software Tools for Performing PCA
    • Summary

Practical Info

Practical Info:

Recommended Duration: 1 day

Course materials :

  • Course notes
  • Sample datasets
  • Course Reviews

    • posted by Joni Keith

      Michel's class was a wonderful introduction to the power of PCA. I was especially impressed with the newfound ability to organize and present sensory data. Given the large number of measurements that are usually taken with sensory testing, this method provides a simple, visual method that allow us to answer so many of the questions we usually ask about the data.

    • posted by Gloria Gaskin

      I enjoyed the PCA class. It was helpful to use some data that we had in house. Michel is very knowledgable in this area. I am anxious to start using it.

    • posted by RD Reeleder

      This course was excellent value for the money. I took this course and the Intro to Cluster Analysis on consecutive days. Both were well-structured and with plenty of hands-on opportunities; it is suited to both beginners and to those with some experience in the technique. The instructors were familiar with all the software packages used by the students and were able to offer practical advice on getting the desired output. A very practical course; loaded with information I could put to use right away. These two courses were followed by a one-day workshop where students were able to work with their own data. Highly recommended.

    • posted by Ping Qiu

      This course is very well structured and instructed. I attended both the PCA and cluster analysis session followed by workshop. The instructor (Natalie) is very knowledgeable and very good at explaining difficult statistical problem in a simple way. This course is especially suitable for non-statistician who needs to perform hands on data analysis. This course also exposed students to many different popular statistics packages so you can get a flavor of each of them which helps me a lot in choosing tools in my future research.

    Related Sessions

    • An applied set of modules with focus on the most widely used multivariate methods and their applications in several fields of application. Learn about the principle of the methods, the data needed, and the information they provide.

    • Learn about preference mapping techniques to explore and understand consumer preferences. Applications dealing with segmentation and the identification of niche markets are discussed. Focus on pitfalls and good practices.

    • Understanding consumers and their preferences is crucial for competitive businesses. Learn about commonly used methods to analyse consumer test data, the way to interpret results and to communicate them with insightful graphical summaries.

    • This workshop on Measuring and Reporting Descriptive Panel Performance first introduces indicators designed to capture the desirable attributes of descriptive data collected by a trained sensory panel (agreement, repeatability and discrimination power). A suite of statistical tools are used to depict and quantify the performance of the panel.

      Real-life case studies are used to illustrate the principles of the statistical tools and the insights that may be extracted from the data.