GraphMath

QRF & equations

Using QR factorization to solve consistent and inconsistent systems

Why solve with QR instead of working directly with AᵀA?

QR factorization separates the transformation by A into an orthogonal part Q and an upper-triangular part R. For a consistent system Ax = b, left-multiplying by Qᵀ removes the rotational part and reduces the problem to solving Rx = Qᵀb. For inconsistent systems, QR gives the same best solution as normal equations while preserving the geometry more cleanly.

Key ideas

QR factorization turns a general system into one whose essential deformation is encoded in an upper-triangular matrix.

  • For a consistent system Ax = b, factoring A = QR once lets us solve many right-hand sides efficiently through Rx = Qᵀb
  • Left-multiplying by Qᵀ removes the orthogonal part of the transformation without requiring inversion of Q
  • In the inconsistent case, QR leads to the same best solution as normal equations but avoids explicitly forming AᵀA
  • Row reduction and QR use different tools and preserve different structures: row reduction targets rows, while QR targets orthogonal columns
  • Back-substitution in R is computationally simpler and geometrically cleaner than solving through AᵀA

The chapter compares these viewpoints directly and shows how the QR method isolates distortion in R while keeping the orthogonal part separate.

What does Qᵀ do when solving Ax = b?

It removes the orthogonal part of the transformation by A. Since Q preserves lengths and angles, multiplying by Qᵀ undoes the rotation or reflection and leaves only the triangular deformation R to solve.

Related chapters

Chapter contents

The chapter is available as a PDF. Page links below are best-effort: most browsers support them but some viewers may ignore the page hint.

Topic Pages
Utility for solving a consistent system A x = b 1
QRF vs row reduction 2
Utility for solving an inconsistent system A x = b 3–6
Normal equations vs QR method 6–7

Why is QR preferred over normal equations?

Because QR solves the same least-squares problem without explicitly forming AᵀA. It keeps the orthogonal part separate, reduces the problem to back-substitution in R and avoids distorting both sides of the equation as much as the normal-equations route does.

Was this chapter helpful?

Quick feedback helps us improve the site.