Concept Architecture
Concept
Theoretically, QR Decomposition is a matrix factorisation method that decomposes a matrix into the product of an orthogonal matrix and an upper triangular matrix. It is a fundamental technique in numerical linear algebra and scientific computing that provides stable algorithms for solving least-squares problems, linear systems and eigenvalue computations. The concept is widely used in health economics for regression estimation, optimisation and numerical modelling.
Mathematically, QR Decomposition expresses a matrix as the product of an orthogonal matrix, whose columns are mutually orthonormal, and an upper triangular matrix. The decomposition may be obtained using methods such as Gram-Schmidt orthogonalisation, Householder reflections or Givens rotations, each producing numerically stable factorisations suitable for statistical computation.
In practice, QR Decomposition is extensively applied in health economic modelling and biostatistics to estimate regression coefficients, solve overdetermined systems of equations, improve numerical stability and support optimisation algorithms. It is the preferred computational method for many regression procedures because it avoids the numerical instability associated with directly inverting cross-product matrices.
Purpose
Used to factorise matrices, solve least-squares problems, estimate regression models, compute eigenvalues and support stable numerical algorithms within health economics.
Mathematical Formulae
Primary Formula
A = QR
where:
- A = original matrix
- Q = orthogonal matrix
- R = upper triangular matrix
Supporting Formulae
Orthogonality Condition
Q?Q = I
Least-Squares Solution
Rx = Q?b
Related Mathematical Methods
- Gram-Schmidt orthogonalisation
- Householder transformation
- Givens rotation
- LU decomposition
- Cholesky decomposition
- Singular value decomposition
Example
Consider the matrix
A =
[ \begin{bmatrix} 1 & 1 \ 1 & -1 \end{bmatrix} ]
Its QR decomposition is
Q =
[ \frac{1}{\sqrt{2}} \begin{bmatrix} 1 & 1 \ 1 & -1 \end{bmatrix} ]
R =
[ \begin{bmatrix} \sqrt{2} & 0 \ 0 & \sqrt{2} \end{bmatrix} ]
Verification:
QR = A
The decomposition therefore reproduces the original matrix while separating its orthogonal and upper triangular components.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| MMULT | =MMULT(Q2:R3,S2:T3) | Verify QR matrix reconstruction. |
| TRANSPOSE | =TRANSPOSE(Q2:R3) | Calculate Q? for least-squares estimation. |
| MUNIT | =MUNIT(2) | Verify orthogonality of Q. |
| MINVERSE | =MINVERSE(A2:B3) | Support related matrix computations. |
| LET/LAMBDA | Custom matrix algorithms | Implement QR decomposition procedures in Excel. |
VBA (Optional)
Automate QR decomposition for regression estimation, optimisation algorithms, eigenvalue computation and numerical methods used in health economic models.
Sources
- Golub GH, Van Loan CF. Matrix Computations.
- Trefethen LN, Bau D. Numerical Linear Algebra.
- Higham NJ. Accuracy and Stability of Numerical Algorithms.
- Press WH, Teukolsky SA, Vetterling WT, Flannery BP. Numerical Recipes.
- Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation.
- Drummond MF, Sculpher MJ, Claxton K, Stoddart GL, Torrance GW. Methods for the Economic Evaluation of Health Care Programmes.
Related Concepts (3)
Library
Publications
1
Accuracy and Stability of Numerical Algorithms — Nicholas J. Higham, 2nd Edition ed., 2002 (Society for Industrial and Applied Mathematics (SIAM))
The standard reference on floating-point error, conditioning, stability and the numerical reliability of matrix, polynomial and linear-algebra algorithms.
BookView source →
Frequently Asked Questions (6)
What is QR decomposition?
A matrix factorisation method that expresses a matrix as the product of an orthogonal matrix and an upper triangular matrix.
Source: Golub GH, Van Loan CF. Matrix Computations. 4th ed. Johns Hopkins University Press; 2013.
What does QR decomposition express a matrix as?
QR decomposition expresses a matrix as the product of an orthogonal matrix and an upper triangular matrix. The original matrix is split into these two factors, one orthogonal and one upper triangular, whose product reconstructs it. Writing a matrix as an orthogonal times an upper triangular matrix is what QR decomposition does Because the orthogonal factor preserves lengths and is numerically well behaved, this factorisation is a stable basis for tasks such as solving least squares problems and computing eigenvalues.
Source: Golub GH, Van Loan CF. Matrix Computations. 4th ed. Johns Hopkins University Press; 2013.
What are the two factors in QR decomposition?
The two factors in QR decomposition are an orthogonal matrix and an upper triangular matrix. The orthogonal factor has columns that are mutually perpendicular and of unit length, while the upper triangular factor holds its non-zero entries on and above the diagonal. These two factors, whose product is the original matrix, are what QR decomposition produces Because the orthogonal factor preserves lengths and is numerically well behaved, this factorisation is a stable basis for tasks such as solving least squares problems and computing eigenvalues.
Source: Golub GH, Van Loan CF. Matrix Computations. 4th ed. Johns Hopkins University Press; 2013.
What is special about the orthogonal factor in QR decomposition?
The orthogonal factor in QR decomposition is special because its columns are mutually perpendicular unit vectors, which gives it numerically convenient properties such as preserving lengths. This makes the factorisation well behaved for computation. Producing such an orthogonal matrix alongside an upper triangular one is a defining feature of QR decomposition Because the orthogonal factor preserves lengths and is numerically well behaved, this factorisation is a stable basis for tasks such as solving least squares problems and computing eigenvalues.
Source: Golub GH, Van Loan CF. Matrix Computations. 4th ed. Johns Hopkins University Press; 2013.
What kind of method is QR decomposition?
QR decomposition is a matrix factorisation method, one that splits a matrix into the product of an orthogonal matrix and an upper triangular matrix. It rewrites the matrix as two structured factors rather than solving a specific numerical problem directly. Being a factorisation into an orthogonal and an upper triangular matrix is what QR decomposition is Because the orthogonal factor preserves lengths and is numerically well behaved, this factorisation is a stable basis for tasks such as solving least squares problems and computing eigenvalues.
Source: Golub GH, Van Loan CF. Matrix Computations. 4th ed. Johns Hopkins University Press; 2013.
How does QR decomposition relate to LU decomposition?
QR decomposition expresses a matrix as the product of an orthogonal matrix and an upper triangular matrix, while LU decomposition expresses a square matrix as the product of a lower triangular matrix and an upper triangular matrix. Both are matrix factorisation methods, differing in the kinds of factors they produce. The two are connected as factorisations that rewrite a matrix into structured factors for easier computation Because the orthogonal factor preserves lengths and is numerically well behaved, this factorisation is a stable basis for tasks such as solving least squares problems and computing eigenvalues.
Source: Golub GH, Van Loan CF. Matrix Computations. 4th ed. Johns Hopkins University Press; 2013.
Trust Record
Verified by Dr Darrin Baines
British health economist
Professional identity: darrinbaines.org
Verification date: 26 May 2026
Content version: 1.0.0
Canonical Identity
- Persistent URI
- https://healtheconomics.wiki/concept/qr-decomposition
- Term code
- CS-LA-MF-002
Stable URI · Machine-readable · Resolvable · CC BY 4.0