Source

Adapted from my own handwritten and hand-typed notes from Year 1 to Year 3 of my EEE degree

Rank and nullity count the dimensions of the subspaces attached to a matrix.

Definitions

For an matrix :

  • , which equals the number of pivots (nonzero rows) in RE form
  • , which equals the number of free variables in

The row space and column space always have the same dimension, so .

Rank Theorem

Reason: in the RRE form of there are pivot (leading) variables and free variables, and each free variable gives one basis vector of the null space.

Properties of Rank

Properties:

  • if exists, and if exists

Worked Example

Two nonzero rows, so . Since , the rank theorem gives . This matches counting directly: 2 independent equations in 4 unknowns leaves 2 free variables.

Rank and Solutions

Rank also decides the solution type of :

  • means no solution
  • means a unique solution
  • means infinitely many solutions

For a square matrix, full rank ties together many conditions, listed in Invertible Matrix Theorem.