Updated: 13 May 2026
To understand how PCA works, we first need to review a few key mathematical concepts. This theory section highlights concepts from linear algebra and statistics that form the building blocks for describing and manipulating data.
Linear Algebra¶
Linear algebra is the language for representing and transforming data in Machine Learning. Since datasets are represented as vectors and matrices, understanding how these objects behave is essential for understanding PCA.
Statistics¶
Statistics provides the tools to describe and understand data. Since PCA is based on identifying patterns in how data varies, we need basic statistical concepts to measure and interpret this variation.
We begin this mathematical review section by revisiting the key concepts from linear algebra.