← 📐 Rank, Elimination & LU
Subtopic 3 of 4

3.3Rank–nullity theorem

The idea

The rank–nullity theorem says that the rank plus the number of free variables equals the number of columns: nothing is lost or gained.

Think of it like this: You have n directions to start with; r survive, and the other n − r are crushed to zero.

Why it exists

It tells you instantly how many independent solutions Ax = 0 has.

How it works

  1. The columns split into pivot columns (r of them) and free columns (n − r).
  2. nullity = n − rank.
  3. So a wide matrix (n > m) always has nullity ≥ n − m > 0.
  4. Each free variable contributes one basis vector of the null space.

Worked example: A 3 × 4 matrix of rank 2

There are 4 columns, so the null space has dimension 4 − 2 = 2: two free variables, two basis vectors.

Now play with it

  • →Watch the bar split into pivot columns and free variables.
  • →Try the 4 × 3 example: rank 3 means the null space is just {0}.
A (3×4), pivot columns shaded
1211
2402
3613
Reduced row echelon form
1201
0010
0000
rank r2
nullity = n − r4 − 2 = 2
dim column space2
dim row space2
dim null space2
dim left null space (m − r)1

The n = 4 columns split into r = 2 (gold: pivot columns) + 2 (green: free variables).

Basis of the column space (pivot columns of A)
1
2
3
1
0
1
Basis of the row space (non-zero rows of the RREF)
1
2
0
1
0
0
1
0
Basis of the null space (solutions of Ax = 0)
-2
1
0
0
-1
0
0
1

Check yourself

A is 5 × 8 with rank 3. The dimension of its null space is

Remember: rank + nullity = number of columns.
Exam notes (the dense version)

For an m×n matrix, rank(A) + nullity(A) = n, where nullity = dimension of the null space {x : Ax = 0}. Free variables in the RREF = nullity. So a wide matrix (more columns than rows) always has a non-trivial null space.

Previous year questions: Rank, Elimination & LU (4)

PYQs are grouped by topic, so this is the full set for Rank, Elimination & LU; try the ones that match this subtopic. Open all 4 on the topic page.