Given by Kristjan Korjus
Brief summary: I would like you to know what is PCA, how and when to use it, how to explain it to someone not familiar with the topic, how to plot and interpret the results of the analysis and how to write a nice report using the method. To sum up, I would like you to be able to get a full skillset needed to use PCA in practice.

This lecture describes the mathematical background of the PCA. Your goal is to apply PCA to the data of the richest and the biggest countries in the world and write a nice report explaining your analysis of the data.

Materials: Lecture notes, R code, Exercise, Data

Literature:

• Bishop: Pattern Recognition and Machine Learning pages 559 - 599

Complementary exercises:

• Bishop: Pattern Recognition and Machine Learning pages 599 - 603
• Try PCA and ICA methods on various image collections and interpret the resullts
  • Semeion Handwritten Digit Data Set
  • Paintings by Edvard Munch
  • Paintings by Jackson Pollock
  • Paintings by Kasimir Malevich

Free implementations:

• Built-in stats package in R: princomp, prcomp