In linear algebra, a QR decomposition, also known as a QR factorization or QU factorization, is a decomposition of a matrix A into a product A = QR of an orthonormal matrix Q and an upper triangular matrix R. QR decomposition is often used to solve the linear least squares (LLS) problem and is the basis for a particular eigenvalue algorithm, the QR algorithm.
Any real square matrix A may be decomposed as
where Q is an orthogonal matrix (its columns are orthogonal unit vectors meaning and R is an upper triangular matrix (also called right triangular matrix). If A is invertible, then the factorization is unique if we require the diagonal elements of R to be positive.
If instead A is a complex square matrix, then there is a decomposition A = QR where Q is a unitary matrix (so the conjugate transpose
If A has n linearly independent columns, then the first n columns of Q form an orthonormal basis for the column space of A. More generally, the first k columns of Q form an orthonormal basis for the span of the first k columns of A for any . The fact that any column k of A only depends on the first k columns of Q corresponds to the triangular form of R.
More generally, we can factor a complex mÃÂn matrix A, with , as the product of an mÃÂm unitary matrix Q and an mÃÂn upper triangular matrix R. As the bottom (mâÂÂn) rows of an mÃÂn upper triangular matrix consist entirely of zeroes, it is often useful to partition R, or both R and Q:
where R<sub>1</sub> is an nÃÂn upper triangular matrix, 0 is an zero matrix, Q<sub>1</sub> is mÃÂn, Q<sub>2</sub> is , and Q<sub>1</sub> and Q<sub>2</sub> both have orthogonal columns.
call Q<sub>1</sub>R<sub>1</sub> the thin QR factorization of A; Trefethen and Bau call this the reduced QR factorization. If A is of full rank n and we require that the diagonal elements of R<sub>1</sub> are positive then R<sub>1</sub> and Q<sub>1</sub> are unique, but in general Q<sub>2</sub> is not. R<sub>1</sub> is then equal to the upper triangular factor of the Cholesky decomposition of A A (= A<sup>T</sup>A if A is real).
Analogously, we can define QL, RQ, and LQ decompositions, with L being a lower triangular matrix.
There are several methods for actually computing the QR decomposition, such as the GramâÂÂSchmidt process, Householder transformations, or Givens rotations. Each has a number of advantages and disadvantages.
Consider the GramâÂÂSchmidt process applied to the columns of the full column rank matrix with inner product (or for the complex case).
Define the projection:
then:
We can now express the s over our newly computed orthonormal basis:
where This can be written in matrix form:
where:
and
Consider the decomposition of
Recall that an orthonormal matrix has the property
Then, we can calculate by means of GramâÂÂSchmidt as follows:
Thus, we have
The RQ decomposition transforms a matrix A into the product of an upper triangular matrix R (also known as right-triangular) and an orthogonal matrix Q. The only difference from QR decomposition is the order of these matrices.
QR decomposition is GramâÂÂSchmidt orthogonalization of columns of A, started from the first column.
RQ decomposition is GramâÂÂSchmidt orthogonalization of rows of A, started from the last row.
The Gram-Schmidt process is inherently numerically unstable. While the application of the projections has an appealing geometric analogy to orthogonalization, the orthogonalization itself is prone to numerical error. A significant advantage is the ease of implementation.
A Householder reflection (or Householder transformation) is a transformation that takes a vector and reflects it about some plane or hyperplane. We can use this operation to calculate the QR factorization of an m-by-n matrix with .
Q can be used to reflect a vector in such a way that all coordinates but one disappear.
Let be an arbitrary real m-dimensional column vector of such that for a scalar ñ. ñ should get the same sign as the -th coordinate of where is to be the pivot coordinate after which all entries are 0 in matrix As final upper triangular form. If the algorithm is implemented using floating-point arithmetic, then ñ should get the opposite sign to avoid loss of significance (for example, when is almost collinear with , becomes "small" and is numerically unstable; the extreme case is , which causes the previous division to result in NaN).
In the complex case, set
and substitute transposition by conjugate transposition in the construction of Q below.
Then, where is the vector , is the Euclidean norm and is an identity matrix, set
Or, if is complex
is an m-by-m Householder matrix, which is both symmetric and orthogonal (Hermitian and unitary in the complex case), and
This can be used to gradually transform an m-by-n matrix A to upper triangular form. First, we multiply A with the Householder matrix Q<sub>1</sub> we obtain when we choose the first matrix column for x. This results in a matrix Q<sub>1</sub>A with zeros in the left column (except for the first row).
This can be repeated for Aâ² (obtained from Q<sub>1</sub>A by deleting the first row and first column), resulting in a Householder matrix Qâ²<sub>2</sub>. Note that Qâ²<sub>2</sub> is smaller than Q<sub>1</sub>. Since we want it really to operate on Q<sub>1</sub>A instead of Aâ² we need to expand it to the upper left, filling in a 1, or in general:
After iterations of this process,
is an upper triangular matrix. So, with
is a QR decomposition of .
This method has greater numerical stability than the GramâÂÂSchmidt method above.
In numerical tests the computed factors and satisfy
at machine precision. Also, orthogonality is preserved: . However, the accuracy of and decrease with condition number:
For a well-conditioned example (, ):
In an ill-conditioned test (, ):
The following table gives the number of operations in the k-th step of the QR-decomposition by the Householder transformation, assuming a square matrix with size n.
Summing these numbers over the steps (for a square matrix of size n), the complexity of the algorithm (in terms of floating point multiplications) is given by
Let us calculate the decomposition of
First, we need to find a reflection that transforms the first column of matrix A, vector into
Now,
and
Here,
Therefore
Now observe:
so we already have almost a triangular matrix. We only need to zero the (3, 2) entry.
Take the (1, 1) minor, and then apply the process again to
By the same method as above, we obtain the matrix of the Householder transformation
after performing a direct sum with 1 to make sure the next step in the process works properly.
Now, we find
Or, to four decimal digits,
The matrix Q is orthogonal and R is upper triangular, so is the required QR decomposition.
The use of Householder transformations is inherently the most simple of the numerically stable QR decomposition algorithms due to the use of reflections as the mechanism for producing zeroes in the R matrix. However, the Householder reflection algorithm is bandwidth heavy and difficult to parallelize, as every reflection that produces a new zero element changes the entirety of both Q and R matrices.
The Householder QR method can be implemented in parallel with algorithms such as the TSQR algorithm (which stands for Tall Skinny QR). This algorithm can be applied in the case when the matrix A has m >> n. This algorithm uses a binary reduction tree to compute local householder QR decomposition at each node in the forward pass, and re-constitute the Q matrix in the backward pass. The binary tree structure aims at decreasing the amount of communication between processor to increase performance.
QR decompositions can also be computed with a series of Givens rotations. Each rotation zeroes an element in the subdiagonal of the matrix, forming the R matrix. The concatenation of all the Givens rotations forms the orthogonal Q matrix.
In practice, Givens rotations are not actually performed by building a whole matrix and doing a matrix multiplication. A Givens rotation procedure is used instead which does the equivalent of the sparse Givens matrix multiplication, without the extra work of handling the sparse elements. The Givens rotation procedure is useful in situations where only relatively few off-diagonal elements need to be zeroed, and is more easily parallelized than Householder transformations.
Let us calculate the decomposition of
First, we need to form a rotation matrix that will zero the lowermost left element, We form this matrix using the Givens rotation method, and call the matrix . We will first rotate the vector to point along the X axis. This vector has an angle We create the orthogonal Givens rotation matrix, :
And the result of now has a zero in the element.
We can similarly form Givens matrices and which will zero the sub-diagonal elements and forming a triangular matrix The orthogonal matrix is formed from the product of all the Givens matrices Thus, we have and the QR decomposition is
The QR decomposition via Givens rotations is the most involved to implement, as the ordering of the rows required to fully exploit the algorithm is not trivial to determine. However, it has a significant advantage in that each new zero element affects only the row with the element to be zeroed (i) and a row above (j). This makes the Givens rotation algorithm more bandwidth efficient and parallelizable than the Householder reflection technique.
It is possible to compute the QR decomposition in a fast way with the use of fast matrix multiplication algorithms in the time for .
We can use QR decomposition to find the determinant of a square matrix. Suppose a matrix is decomposed as . Then we have
can be chosen such that . Thus,
where the are the entries on the diagonal of . Furthermore, because the determinant equals the product of the eigenvalues, we have
where the are eigenvalues of .
We can extend the above properties to a non-square complex matrix by introducing the definition of QR decomposition for non-square complex matrices and replacing eigenvalues with singular values.
Start with a QR decomposition for a non-square matrix A:
where denotes the zero matrix and is a unitary matrix.
From the properties of the singular value decomposition (SVD) and the determinant of a matrix, we have
where the are the singular values of
Note that the singular values of and are identical, although their complex eigenvalues may be different. However, if A is square, then
It follows that the QR decomposition can be used to efficiently calculate the product of the eigenvalues or singular values of a matrix.
Pivoted QR differs from ordinary Gram-Schmidt in that it takes the largest remaining column at the beginning of each new stepâÂÂcolumn pivotingâ and thus introduces a permutation matrix P:
Column pivoting is useful when A is (nearly) rank deficient, or is suspected of being so. It can also improve numerical accuracy. P is usually chosen so that the diagonal elements of R are non-increasing: . This can be used to find the (numerical) rank of A at lower computational cost than a singular value decomposition, forming the basis of so-called rank-revealing QR algorithms.
Compared to the direct matrix inverse, inverse solutions using QR decomposition are more numerically stable as evidenced by their reduced condition numbers.
To solve the underdetermined linear problem where the matrix has dimensions and rank first find the QR factorization of the transpose of where Q is an orthogonal matrix (i.e. and R has a special form: . Here is a square right triangular matrix, and the zero matrix has dimension After some algebra, it can be shown that a solution to the inverse problem can be expressed as: where one may either find by Gaussian elimination or compute directly by forward substitution. The latter technique enjoys greater numerical accuracy and lower computations.
To find a solution to the overdetermined problem which minimizes the norm first find the QR factorization of The solution can then be expressed as where is an matrix containing the first columns of the full orthonormal basis and where is as before. Equivalent to the underdetermined case, back substitution can be used to quickly and accurately find this without explicitly inverting ( and are often provided by numerical libraries as an "economic" QR decomposition.)
Iwasawa decomposition generalizes QR decomposition to semi-simple Lie groups.