notesonly.in

One notebook for every subject — open it anywhere.

Log in

Transpose of a Matrix: Definition, Properties, Examples

MathematicsTry the interactive 3D model →

The transpose of a matrix A, written A^T (or A′), is formed by turning its rows into columns: the entry in row i, column j of A moves to row j, column i of A^T. So (A^T)_ij = a_ji, and an m × n matrix becomes an n × m matrix.

The transpose matters because it appears in almost every matrix formula you will meet — from (AB)^T = B^TA^T to the definition of symmetric matrices (A^T = A). It converts a row vector into a column vector and back, something linear algebra does constantly.

Key Concepts

  • Definition: (A^T)_ij = a_ji. Rows and columns swap; the main diagonal stays fixed in place.
  • Order flips: if A is m × n, then A^T is n × m.
  • Double transpose: (A^T)^T = A — transposing twice returns the original.
  • Transpose of a sum: (A + B)^T = A^T + B^T.
  • Transpose of a product: (AB)^T = B^TA^T — note the order reverses, like taking off shoes then socks.
  • Scalar: (kA)^T = kA^T.

How It Works

To transpose a matrix:

  1. Write the first row as the first column: the entries a_11, a_12, …, a_1n become the first column from top to bottom.
  2. Repeat for each row: row 2 becomes column 2, row 3 becomes column 3, and so on.
  3. Check the new order: count — you should now have n rows and m columns.
  4. Verify with the diagonal: the entries a_11, a_22, … should sit in exactly the same spots as before; everything else has mirrored across that diagonal.

Worked Example

Find A^T and verify (A^T)^T = A for A = [[1, 2, 3], [4, 5, 6]].

  1. Row 1 → column 1: [1, 2, 3] becomes the first column: 1 on top, then 2, then 3.
  2. Row 2 → column 2: [4, 5, 6] becomes the second column.
  3. Result: A^T = [[1, 4], [2, 5], [3, 6]] — a 3 × 2 matrix.
  4. Transpose again: rows of A^T are [1, 4], [2, 5], [3, 6]; turning them into columns gives [[1, 2, 3], [4, 5, 6]] = A. Verified.
  5. Spot check one entry: a_23 = 6 in A; in A^T it sits at (3,2) = 6. Correct — indices swap.

Common Misconceptions

  • "(AB)^T = A^TB^T." Wrong — the order flips: (AB)^T = B^TA^T. Try it with numbers once and the mistake becomes obvious.
  • "Transpose changes the numbers." It never changes a single value — it only moves entries to new positions. The multiset of entries is identical.
  • "Only square matrices can be transposed." Any matrix of any order can be transposed; the result is just a different shape (n × m).

Real-World Applications

  • Data tables: transposing swaps rows and columns — turning a "students × subjects" marks table into "subjects × students" for subject-wise analysis.
  • Dot products: the dot product of vectors x and y is written x^Ty, the foundation of projections and least-squares fitting.
  • Symmetric matrices: covariance matrices in statistics satisfy A^T = A; the transpose is how symmetry is defined and tested.
  • Computer graphics: the transpose of a rotation matrix is its inverse — a fact graphics engines exploit to undo rotations cheaply.

The transpose swaps rows with columns, turning m × n into n × m while leaving every value untouched. Its key rule is the order-reversing (AB)^T = B^TA^T, behind symmetric matrices and dot products written as x^Ty.

Related blogs