|
|
| Line 1: |
Line 1: |
| In [[linear algebra]], a '''QR decomposition''' (also called a '''QR factorization''') of a [[matrix (mathematics)|matrix]] is a [[matrix decomposition|decomposition]] of a matrix ''A'' into a product ''A'' = ''QR'' of an [[orthogonal matrix]] ''Q'' and an [[upper triangular matrix]] ''R''. QR decomposition is often used to solve the [[linear least squares (mathematics)|linear least squares]] problem, and is the basis for a particular [[eigenvalue algorithm]], the [[QR algorithm]].
| | Types is suffering and our resources are depleting because we use too much of everything. Did you know that taking little steps to save types may also help you save money? If you follow one or all of the ideas below, you can see real savings on your power bill. You don"t have to do every thing but you can do something. In the event people want to be taught supplementary information on [http://well-behaved.info/2014/08/20/how-to-select-your-proper-shower-basket/ go here for more info], we recommend thousands of libraries you could pursue. <br><br>In a normal U.S. home, appliances and electronics make up about two decades of energy use. The first step would be to look for Energy Star labels when buying services and products and appliances. Energy-star labels are strict directions created by the U.S. Department of Energy and Environmental Protection Agency for energy efficiency. Follow several of those ideas to help make the surroundings better for our future generations and save money at once. <br><br>Light bulbs: Use energy-saving CFLs (Compact Fluorescent Light bulbs) as opposed to the traditional incandescent light bulbs. CFLs only use one-quarter of the energy and continues years longer, though CFLs charge 3-5 times up to the incandescent lamp. Which means you could have a supplementary product to sort in the recycling bin each [http://search.Un.org/search?ie=utf8&site=un_org&output=xml_no_dtd&client=UN_Website_en&num=10&lr=lang_en&proxystylesheet=UN_Website_en&oe=utf8&q=CFL+lamp&Submit=Go CFL lamp] contains 5mg of mercury. <br><br>Until they are a complete load dishes: Never load your dishes in a [http://Www.Answers.com/topic/dishwasher dishwasher]. Air-dry meals as opposed to using a dryer. <br><br>Laundry: Wash clothes with hot or cold water instead of hot water when you are able. In place of utilizing a dryer, air-dry or hang your clothes on the clothes line. Don"t make use of the dryer until it houses the full load. Clicking [http://autopilot-income-machines.info/blogs/saving-cash-on-electrical-power-a-funds-saving-maintenance/ Saving Cash on Electrical power: A Funds-Saving Maintenance | Auto Pilot Income Machi] seemingly provides warnings you might tell your aunt. <br><br>Refrigerators: Consider what you want to sign up for before you open the fridge. Dig up more on this partner wiki - Click here: [http://running55.info/2014/08/20/how-to-decide-on-your-correct-shower-basket/ pipe lining inspection in burbank]. Leaving the door open will waste extra power. <br><br>Take Showers, less Baths: Taking baths instead of baths will also decrease your heating bill and reduce water usage. <br><br>Turn Off all Appliances Not in Use: Turn off all electronics, computers, and lights if they aren"t being used. Select all your technology including computers, TVs, and DVDs in-to power strips. Using a power strip could reduce energy used to power appliances for the home. Even though your devices are switched off, power remains being consumed from the outlet. To prevent extra energy costs, disconnect the applying or work with a power strip. The power strip comes with an off switch to reduce all the power from your equipment. <br><br>Weather-Strip Windows and Doors: Always check windows and doors for air leaks. Air leaks could be made by caulking or weather-stripping. By acquiring the leaks in your home, warm and cool air is going to be held in your home longer. Less heating and cooling will help you save on energy prices. <br><br>Try to find Energy Star label: Energy Star labels are merely provided when appliances and products meet strict guidelines of energy efficiency from U.S. Department of Energy and Environmental Protection Agency..<br><br>If you have any concerns about where by and how to use health america - [http://voraciousrainbo88.jimdo.com click through the next article],, you can contact us at the webpage. |
| | |
| 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 specifically, the first ''k'' columns of ''Q'' form an orthonormal basis for the [[linear span|span]] of the first ''k'' columns of ''A'' for any 1 ≤ ''k'' ≤ ''n''.<ref name=Trefethen>L. N. Trefethen and D. Bau, ''Numerical Linear Algebra'' (SIAM, 1997).</ref> The fact that any column ''k'' of ''A'' only depends on the first ''k'' columns of ''Q'' is responsible for the triangular form of ''R''.<ref name=Trefethen/>
| |
| | |
| == Definitions ==
| |
| | |
| ===Square matrix===
| |
| Any real square matrix ''A'' may be decomposed as
| |
| : <math> A = QR, \, </math>
| |
| where ''Q'' is an [[orthogonal matrix]] (its columns are [[orthogonal]] [[unit vector]]s meaning ''Q''<sup>T</sup>''Q'' = ''I'') and ''R'' is an upper [[triangular matrix]] (also called right triangular matrix). This generalizes to a complex square matrix ''A'' and a [[unitary matrix]] ''Q'' (where ''Q''<sup>*</sup>''Q'' = ''I''). If ''A'' is [[invertible matrix|invertible]], then the factorization is unique if we require that the diagonal elements of ''R'' are positive.
| |
| | |
| ===Rectangular matrix===
| |
| More generally, we can factor a complex ''m''×''n'' matrix ''A'', with ''m'' ≥ ''n'', as the product of an ''m''×''m'' [[unitary matrix]] ''Q'' and an ''m''×''n'' upper triangular matrix ''R''. As the bottom (''m''−''n'') rows of an ''m''×''n'' upper triangular matrix consist entirely of zeroes, it is often useful to partition ''R'', or both ''R'' and ''Q'':
| |
| :<math>
| |
| A = QR = Q \begin{bmatrix} R_1 \\ 0 \end{bmatrix}
| |
| = \begin{bmatrix} Q_1, Q_2 \end{bmatrix} \begin{bmatrix} R_1 \\ 0 \end{bmatrix}
| |
| = Q_1 R_1,
| |
| </math>
| |
| where ''R''<sub>1</sub> is an ''n''×''n'' upper triangular matrix, ''Q''<sub>1</sub> is ''m''×''n'', ''Q''<sub>2</sub> is ''m''×(''m'' − ''n''), and ''Q''<sub>1</sub> and ''Q''<sub>2</sub> both have orthogonal columns.
| |
| | |
| {{harvtxt|Golub|Van Loan|1996|loc=§5.2}} call ''Q''<sub>1</sub>''R''<sub>1</sub> the ''thin QR factorization'' of ''A''; Trefethen and Bau call this the ''reduced QR factorization''.<ref name=Trefethen/>
| |
| If ''A'' is of full [[matrix rank|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''{{starred}} ''A'' (= ''A''<sup>T</sup>''A'' if ''A'' is real).
| |
| | |
| ===QL, RQ and LQ decompositions===
| |
| Analogously, we can define QL, RQ, and LQ decompositions, with ''L'' being a ''lower'' triangular matrix.
| |
| | |
| ==Computing the QR decomposition==
| |
| There are several methods for actually computing the QR decomposition, such as by means of the [[Gram–Schmidt process]], [[Householder transformation]]s, or [[Givens rotation]]s. Each has a number of advantages and disadvantages.
| |
| | |
| === Using the Gram–Schmidt process ===
| |
| {{details|Gram–Schmidt#Numerical stability}}
| |
| Consider the [[Gram–Schmidt process]] applied to the columns of the full column rank matrix <math>A=[\mathbf{a}_1, \cdots, \mathbf{a}_n]</math>, with [[inner product]] <math>\langle\mathbf{v},\mathbf{w}\rangle = \mathbf{v}^\top \mathbf{w}</math> (or <math>\langle\mathbf{v},\mathbf{w}\rangle = \mathbf{v}^* \mathbf{w}</math> for the complex case).
| |
| | |
| Define the [[Vector projection|projection]]:
| |
| :<math>\mathrm{proj}_{\mathbf{e}}\mathbf{a}
| |
| = \frac{\left\langle\mathbf{e},\mathbf{a}\right\rangle}{\left\langle\mathbf{e},\mathbf{e}\right\rangle}\mathbf{e}
| |
| </math>
| |
| then:
| |
| :<math>
| |
| \begin{align}
| |
| \mathbf{u}_1 &= \mathbf{a}_1,
| |
| & \mathbf{e}_1 &= {\mathbf{u}_1 \over \|\mathbf{u}_1\|} \\
| |
| \mathbf{u}_2 &= \mathbf{a}_2-\mathrm{proj}_{\mathbf{e}_1}\,\mathbf{a}_2,
| |
| & \mathbf{e}_2 &= {\mathbf{u}_2 \over \|\mathbf{u}_2\|} \\
| |
| \mathbf{u}_3 &= \mathbf{a}_3-\mathrm{proj}_{\mathbf{e}_1}\,\mathbf{a}_3-\mathrm{proj}_{\mathbf{e}_2}\,\mathbf{a}_3,
| |
| & \mathbf{e}_3 &= {\mathbf{u}_3 \over \|\mathbf{u}_3\|} \\
| |
| & \vdots &&\vdots \\
| |
| \mathbf{u}_k &= \mathbf{a}_k-\sum_{j=1}^{k-1}\mathrm{proj}_{\mathbf{e}_j}\,\mathbf{a}_k,
| |
| &\mathbf{e}_k &= {\mathbf{u}_k\over\|\mathbf{u}_k\|}
| |
| \end{align}
| |
| </math>
| |
| | |
| We then rearrange the equations above so that the <math>\mathbf{a}_i</math>s are on the left, using the fact that the <math>\mathbf{e}_i</math> are unit vectors:
| |
| | |
| :<math> | |
| \begin{align}
| |
| \mathbf{a}_1 &= \langle\mathbf{e}_1,\mathbf{a}_1 \rangle \mathbf{e}_1 \\
| |
| \mathbf{a}_2 &= \langle\mathbf{e}_1,\mathbf{a}_2 \rangle \mathbf{e}_1
| |
| + \langle\mathbf{e}_2,\mathbf{a}_2 \rangle \mathbf{e}_2 \\
| |
| \mathbf{a}_3 &= \langle\mathbf{e}_1,\mathbf{a}_3 \rangle \mathbf{e}_1
| |
| + \langle\mathbf{e}_2,\mathbf{a}_3 \rangle \mathbf{e}_2
| |
| + \langle\mathbf{e}_3,\mathbf{a}_3 \rangle \mathbf{e}_3 \\
| |
| &\vdots \\
| |
| \mathbf{a}_k &= \sum_{j=1}^{k} \langle \mathbf{e}_j, \mathbf{a}_k \rangle \mathbf{e}_j
| |
| \end{align}
| |
| </math>
| |
| where <math>\langle\mathbf{e}_i,\mathbf{a}_i \rangle = \|\mathbf{u}_i\|</math>. This can be written in matrix form:
| |
| :<math> A = Q R </math>
| |
| where:
| |
| :<math>Q = \left[ \mathbf{e}_1, \cdots, \mathbf{e}_n\right] \qquad \text{and} \qquad
| |
| R = \begin{pmatrix}
| |
| \langle\mathbf{e}_1,\mathbf{a}_1\rangle & \langle\mathbf{e}_1,\mathbf{a}_2\rangle & \langle\mathbf{e}_1,\mathbf{a}_3\rangle & \ldots \\
| |
| 0 & \langle\mathbf{e}_2,\mathbf{a}_2\rangle & \langle\mathbf{e}_2,\mathbf{a}_3\rangle & \ldots \\
| |
| 0 & 0 & \langle\mathbf{e}_3,\mathbf{a}_3\rangle & \ldots \\
| |
| \vdots & \vdots & \vdots & \ddots \end{pmatrix}.</math>
| |
| | |
| ==== Example ====
| |
| Consider the decomposition of
| |
| : <math>A =
| |
| \begin{pmatrix}
| |
| 12 & -51 & 4 \\
| |
| 6 & 167 & -68 \\
| |
| -4 & 24 & -41
| |
| \end{pmatrix}
| |
| .</math> | |
| | |
| Recall that an orthogonal matrix <math>Q</math> has the property
| |
| : <math>
| |
| \begin{matrix}
| |
| Q^T \,Q = I.
| |
| \end{matrix}
| |
| </math>
| |
| | |
| Then, we can calculate <math>Q</math> by means of Gram–Schmidt as follows:
| |
| : <math>
| |
| U =
| |
| \begin{pmatrix}
| |
| \mathbf u_1 & \mathbf u_2 & \mathbf u_3
| |
| \end{pmatrix}
| |
| =
| |
| \begin{pmatrix}
| |
| 12 & -69 & -58/5 \\
| |
| 6 & 158 & 6/5 \\
| |
| -4 & 30 & -33 | |
| \end{pmatrix};
| |
| </math>
| |
| : <math>
| |
| Q =
| |
| \begin{pmatrix}
| |
| \frac{\mathbf u_1}{\|\mathbf u_1\|} &
| |
| \frac{\mathbf u_2}{\|\mathbf u_2\|} &
| |
| \frac{\mathbf u_3}{\|\mathbf u_3\|}
| |
| \end{pmatrix}
| |
| =
| |
| \begin{pmatrix}
| |
| 6/7 & -69/175 & -58/175 \\
| |
| 3/7 & 158/175 & 6/175 \\
| |
| -2/7 & 6/35 & -33/35
| |
| \end{pmatrix}.
| |
| </math>
| |
| | |
| Thus, we have
| |
| : <math>
| |
| \begin{matrix}
| |
| Q^{T} A = Q^{T}Q\,R = R;
| |
| \end{matrix}
| |
| </math> | |
| : <math>
| |
| \begin{matrix}
| |
| R = Q^{T}A =
| |
| \end{matrix}
| |
| \begin{pmatrix}
| |
| 14 & 21 & -14 \\
| |
| 0 & 175 & -70 \\
| |
| 0 & 0 & 35
| |
| \end{pmatrix}.
| |
| </math>
| |
| | |
| ====Relation to RQ decomposition====
| |
| 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 Gram–Schmidt orthogonalization of columns of ''A'', started from the first column.
| |
| | |
| RQ decomposition is Gram–Schmidt orthogonalization of rows of ''A'', started from the last row.
| |
| | |
| === Using Householder reflections ===
| |
| [[File:Householder.svg|thumb|Householder reflection for QR-decomposition: The goal is to find a linear transformation that changes the vector <math>x</math> into a vector of same length which is collinear to <math>e_1</math>. We could use an orthogonal projection (Gram-Schmidt) but this will be numerically unstable if the vectors <math>x</math> and <math>e_1</math> are close to orthogonal. Instead, the Householder reflection reflects through the dotted line (chosen to bisect the angle between <math>x</math> and <math>e_1</math>). The maximum angle with this transform is at most 45 degrees.]]
| |
| | |
| A [[Householder reflection]] (or ''Householder transformation'') is a transformation that takes a vector and reflects it about some [[plane (mathematics)|plane]] or [[hyperplane]]. We can use this operation to calculate the ''QR'' factorization of an ''m''-by-''n'' matrix <math>A</math> with ''m'' ≥ ''n''.
| |
| | |
| ''Q'' can be used to reflect a vector in such a way that all coordinates but one disappear.
| |
| | |
| Let <math>\mathbf{x}</math> be an arbitrary real ''m''-dimensional column vector of <math>A</math> such that <math>\|\mathbf{x}\| = |\alpha|</math> for a scalar ''α''. If the algorithm is implemented using [[floating-point arithmetic]], then ''α'' should get the opposite sign as the ''k''-th coordinate of <math>\mathbf{x}</math>, where <math>x_k</math> is to be the pivot coordinate after which all entries are 0 in matrix ''A'''s final upper triangular form, to avoid [[loss of significance]]. In the complex case, set
| |
| :<math> \alpha = - \mathrm{e}^{\mathrm{i} \arg x_k} \|\mathbf{x}\|</math>
| |
| {{harv|Stoer|Bulirsch|2002|p=225}} and substitute transposition by conjugate transposition in the construction of ''Q'' below.
| |
| | |
| Then, where <math>\mathbf{e}_1</math> is the vector (1,0,...,0)<sup>T</sup>, ||·|| is the [[Euclidean space#Euclidean structure|Euclidean norm]] and <math>I</math> is an ''m''-by-''m'' identity matrix, set
| |
| : <math>\mathbf{u} = \mathbf{x} - \alpha\mathbf{e}_1,</math>
| |
| : <math>\mathbf{v} = {\mathbf{u}\over\|\mathbf{u}\|},</math>
| |
| : <math>Q = I - 2 \mathbf{v}\mathbf{v}^T.</math>
| |
| | |
| Or, if <math>A</math> is complex
| |
| : <math>Q = I - (1+w)\mathbf{v}\mathbf{v}^H</math>, where <math>w = \mathbf{x}^H\mathbf{v}\mathbf{/}\mathbf{v}^H\mathbf{x} </math>
| |
| : where <math>\mathbf{x}^H</math> is the [[conjugate transpose]] (transjugate) of <math>\mathbf{x}</math>
| |
| | |
| <math>Q</math> is an ''m''-by-''m'' Householder matrix and
| |
| : <math>Q\mathbf{x} = (\alpha, 0, \cdots, 0)^T.\,</math>
| |
| | |
| 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).
| |
| | |
| :<math>Q_1A = \begin{bmatrix}
| |
| \alpha_1&\star&\dots&\star\\
| |
| 0 & & & \\
| |
| \vdots & & A' & \\
| |
| 0 & & & \end{bmatrix}</math>
| |
| | |
| This can be repeated for ''A''′ (obtained from ''Q''<sub>1</sub>''A'' by deleting the first row and first column), resulting in a Householder matrix ''Q''′<sub>2</sub>. Note that ''Q''′<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 ''A''′ we need to expand it to the upper left, filling in a 1, or in general:
| |
| | |
| :<math>Q_k = \begin{pmatrix}
| |
| I_{k-1} & 0\\
| |
| 0 & Q_k'\end{pmatrix}.</math>
| |
| | |
| After <math>t</math> iterations of this process, <math>t = \min(m-1, n)</math>,
| |
| :<math> R = Q_t \cdots Q_2Q_1A</math> | |
| is an upper triangular matrix. So, with
| |
| :<math> Q = Q_1^T Q_2^T \cdots Q_t^T, </math>
| |
| <math>A = QR</math> is a QR decomposition of <math>A</math>.
| |
| | |
| This method has greater [[numerical stability]] than the Gram–Schmidt method above. <!--See the below example, and compare above-->
| |
| | |
| 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''.
| |
| {| class="wikitable"
| |
| |-
| |
| ! Operation
| |
| ! Number of operations in the ''k''-th step
| |
| |-
| |
| | multiplications
| |
| | <math>2(n-k+1)^2</math>
| |
| |-
| |
| | additions
| |
| | <math> (n-k+1)^2+(n-k+1)(n-k)+2 </math>
| |
| |-
| |
| | division
| |
| | <math>1</math>
| |
| |-
| |
| | square root
| |
| | <math>1</math>
| |
| |}
| |
| | |
| Summing these numbers over the ''n'' − 1 steps (for a square matrix of size ''n''), the complexity of the algorithm (in terms of floating point multiplications) is given by
| |
| :<math>\frac{2}{3}n^3+n^2+\frac{1}{3}n-2=O(n^3).</math>
| |
| | |
| ==== Example ====
| |
| Let us calculate the decomposition of
| |
| : <math>A = \begin{pmatrix}
| |
| 12 & -51 & 4 \\
| |
| 6 & 167 & -68 \\
| |
| -4 & 24 & -41 \end{pmatrix}.</math>
| |
| | |
| First, we need to find a reflection that transforms the first column of matrix ''A'', vector <math>\mathbf{a}_1 = (12, 6, -4)^T</math>, to <math>\|\mathbf{a}_1\| \;\mathrm{e}_1 = (14, 0, 0)^T.</math>
| |
| | |
| Now,
| |
| : <math>\mathbf{u} = \mathbf{x} + \alpha\mathbf{e}_1,</math> | |
| and
| |
| : <math>\mathbf{v} = {\mathbf{u}\over\|\mathbf{u}\|}.</math>
| |
| Here,
| |
| : <math>\alpha = -14</math> and <math>\mathbf{x} = \mathbf{a}_1 = (12, 6, -4)^T</math>
| |
| | |
| Therefore
| |
| : <math>\mathbf{u} = (-2, 6, -4)^T=({2})(-1, 3, -2)^T</math> and <math>\mathbf{v} = {1 \over \sqrt{14}}(-1, 3, -2)^T</math>, and then
| |
| :<math>Q_1 = I - {2 \over \sqrt{14} \sqrt{14}} \begin{pmatrix} -1 \\ 3 \\ -2 \end{pmatrix}\begin{pmatrix} -1 & 3 & -2 \end{pmatrix}</math>
| |
| :<math> = I - {1 \over 7}\begin{pmatrix}
| |
| 1 & -3 & 2 \\
| |
| -3 & 9 & -6 \\
| |
| 2 & -6 & 4
| |
| \end{pmatrix}</math>
| |
| :<math> = \begin{pmatrix} | |
| 6/7 & 3/7 & -2/7 \\
| |
| 3/7 &-2/7 & 6/7 \\
| |
| -2/7 & 6/7 & 3/7 \\
| |
| \end{pmatrix}.</math>
| |
| | |
| Now observe:
| |
| :<math>Q_1A = \begin{pmatrix}
| |
| 14 & 21 & -14 \\
| |
| 0 & -49 & -14 \\
| |
| 0 & 168 & -77 \end{pmatrix},</math>
| |
| so we already have almost a triangular matrix. We only need to zero the (3, 2) entry.
| |
| | |
| Take the (1, 1) [[minor (linear algebra)|minor]], and then apply the process again to
| |
| :<math>A' = M_{11} = \begin{pmatrix} | |
| -49 & -14 \\
| |
| 168 & -77 \end{pmatrix}.</math>
| |
| By the same method as above, we obtain the matrix of the Householder transformation
| |
| :<math>Q_2 = \begin{pmatrix}
| |
| 1 & 0 & 0 \\
| |
| 0 & 7/25 & -24/25 \\
| |
| 0 & -24/25 & -7/25 \end{pmatrix}</math>
| |
| after performing a direct sum with 1 to make sure the next step in the process works properly.
| |
| | |
| Now, we find
| |
| :<math>Q=Q_1^T Q_2^T=\begin{pmatrix}
| |
| 6/7 & 69/175 & -58/175 \\
| |
| 3/7 & -158/175 & 6/175 \\
| |
| -2/7 & -6/35 & -33/35 \end{pmatrix} </math> | |
| Then
| |
| :<math>Q=Q_1^T Q_2^T=\begin{pmatrix}
| |
| 0.8571 & 0.3943 & -0.3314 \\
| |
| 0.4286 & -0.9029 & 0.0343 \\
| |
| -0.2857 & -0.1714 & -0.9429 \end{pmatrix} </math> | |
| | |
| :<math>R=Q_2Q_1A=Q^T A=\begin{pmatrix}
| |
| 14 & 21 & -14 \\
| |
| 0 & -175 & 70 \\
| |
| 0 & 0 & 35 \end{pmatrix}.</math>
| |
| The matrix ''Q'' is orthogonal and ''R'' is upper triangular, so ''A'' = ''QR'' is the required QR-decomposition.
| |
| | |
| === Using Givens rotations ===
| |
| ''QR'' decompositions can also be computed with a series of [[Givens rotation]]s. Each rotation zeros 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 a relatively few off diagonal elements need to be zeroed, and is more easily parallelized than [[Householder transformation]]s.
| |
| | |
| ==== Example ====
| |
| Let us calculate the decomposition of
| |
| : <math>A = \begin{pmatrix}
| |
| 12 & -51 & 4 \\
| |
| 6 & 167 & -68 \\
| |
| -4 & 24 & -41 \end{pmatrix}.</math>
| |
| | |
| First, we need to form a rotation matrix that will zero the lowermost left element, <math>\mathbf{a}_{31} = -4</math>. We form this matrix using the Givens rotation method, and call the matrix <math>G_1</math>. We will first rotate the vector <math>(6,-4)</math>, to point along the ''X'' axis. This vector has an angle <math>\theta = \arctan\left({-(-4) \over 6}\right)</math>. We create the orthogonal Givens rotation matrix, <math>G_1</math>:
| |
| | |
| :<math>G_1 = \begin{pmatrix}
| |
| 1 & 0 & 0 \\
| |
| 0 & \cos(\theta) & -\sin(\theta) \\
| |
| 0 & \sin(\theta) & \cos(\theta)
| |
| \end{pmatrix}</math>
| |
| :<math>\approx \begin{pmatrix}
| |
| 1 & 0 & 0 \\
| |
| 0 & 0.83205 & -0.55470 \\
| |
| 0 & 0.55470 & 0.83205
| |
| \end{pmatrix}</math>
| |
| | |
| And the result of <math>G_1A</math> now has a zero in the <math>\mathbf{a}_{31}</math> element.
| |
| :<math>G_1A \approx \begin{pmatrix} | |
| 12 & -51 & 4 \\
| |
| 7.21110 & 125.6396 & -33.83671 \\
| |
| 0 & 112.6041 & -71.83368
| |
| \end{pmatrix}</math>
| |
| | |
| We can similarly form Givens matrices <math>G_2</math> and <math>G_3</math>, which will zero the sub-diagonal elements <math>a_{21}</math> and <math>a_{32}</math>, forming a triangular matrix <math>R</math>. The orthogonal matrix <math>Q^T</math> is formed from the concatenation of all the Givens matrices <math>Q^T = G_3G_2G_1</math>. Thus, we have <math>
| |
| G_3G_2G_1A= Q^TA = R</math>, and the ''QR'' decomposition is <math>A = QR</math>.
| |
| | |
| ==Connection to a determinant or a product of eigenvalues==
| |
| We can use QR decomposition to find the absolute value of the [[determinant]] of a square matrix. Suppose a matrix is decomposed as <math>A=QR</math>. Then we have
| |
| :<math>\det(A)=\det(Q)\cdot\det(R).</math>
| |
| Since ''Q'' is unitary, <math>|\det(Q)|=1</math>. Thus,
| |
| :<math>|\det(A)|=|\det(R)|=\Big|\prod_{i} r_{ii}\Big|,</math>
| |
| where <math>r_{ii}</math> are the entries on the diagonal of ''R''.
| |
| | |
| Furthermore, because the determinant equals the product of the eigenvalues, we have
| |
| :<math>\Big|\prod_{i} r_{ii}\Big|=\Big|\prod_{i} \lambda_{i}\Big|,</math>
| |
| where <math>\lambda_{i}</math> are eigenvalues of <math>A</math>.
| |
| | |
| We can extend the above properties to non-square complex matrix <math>A</math> by introducing the definition of QR-decomposition for non-square complex matrix and replacing eigenvalues with singular values.
| |
| | |
| Suppose a QR decomposition for a non-square matrix ''A'':
| |
| :<math>A = Q \begin{pmatrix}R\\O\end{pmatrix}, \qquad Q^*Q = I,</math>
| |
| where <math>O</math> is a zero matrix and <math>Q</math> is a unitary matrix.
| |
| | |
| From the properties of [[Singular value decomposition|SVD]] and determinant of matrix, we have
| |
| :<math>\Big|\prod_{i} r_{ii}\Big| = \prod_{i} \sigma_{i},</math>
| |
| where <math>\sigma_{i}</math> are singular values of <math>A</math>.
| |
| | |
| Note that the singular values of <math>A</math> and <math>R</math> are identical, although their complex eigenvalues may be different. However, if ''A'' is square, the following is true:
| |
| :<math>
| |
| {\prod_{i} \sigma_{i}} = \Big|{\prod_{i} \lambda_{i}}\Big|.
| |
| </math>
| |
| | |
| In conclusion, QR decomposition can be used efficiently to calculate the product of the eigenvalues or singular values of a matrix.
| |
| | |
| ==Column pivoting==
| |
| {{Expand section|date=December 2009}}
| |
| QR decomposition with column pivoting introduces a [[permutation matrix]] ''P'':
| |
| :<math>A P = Q R \quad \iff A = Q R P^T </math>
| |
| 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:
| |
| <math>|r_{11}| \ge |r_{22}| \ge \ldots \ge |r_{nn}|</math>. 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 algorithm]]s.
| |
| | |
| ==Using for solution to linear inverse problems==
| |
| Compared to the direct matrix inverse, inverse solutions using QR decomposition are more numerically stable as evidenced by their reduced [[condition number]]s [Parker, Geophysical Inverse Theory, Ch1.13].
| |
| | |
| To solve the underdetermined(<math>m < n </math>) linear problem <math> Ax=b </math> where the matrix A has dimensions <math>m \times n</math> and rank <math>m</math>, first find the QR factorization of the transpose of A: <math> A^T=QR </math>, where Q is an orthogonal matrix (i.e. <math>Q^T=Q^{-1}</math>), and R has a special form: <math>R=\begin{bmatrix} R_1 \\ 0\end{bmatrix}</math>. Here <math>R_1</math> is a square <math>m \times m</math> right triangular matrix, and the zero matrix has dimension <math>(n-m) \times m)</math>. After some algebra, it can be shown that a solution to the inverse problem can be expressed as: <math>
| |
| x = Q
| |
| \begin{bmatrix}
| |
| (R_1^T)^{-1}b \\
| |
| 0
| |
| \end{bmatrix}
| |
| </math>
| |
| where one may either find <math>R_1^{-1}</math> by Gaussian elimination or compute <math>(R_1^T)^{-1} b</math> directly by [[Triangular matrix#Forward and back_substitution|forward substitution]]. The latter technique enjoys greater numerical accuracy and lower computations.
| |
| | |
| To find a solution, <math>\hat x</math>, to the overdetermined (<math> m \geq n </math>) problem <math>Ax=b</math> which minimizes the norm <math>\|A \hat x-b\|</math>, first find the QR factorization of A: <math>A=QR</math>. The solution can then be expressed as <math> \hat x = R_1^{-1} (Q_1^T b) </math>, where <math>Q_1</math> is an <math>m \times n</math> matrix containing the first <math>n</math> columns of the full orthonormal basis <math>Q</math> and where <math>R_1</math> is as before. Equivalent to the underdetermined case, [[Triangular matrix#Forward and back_substitution|back substitution]] can be used to quickly and accurately find this <math>\hat x</math> without explicitly inverting <math>R_1</math>. (<math>Q_1</math> and <math>R_1</math> are often provided by numerical libraries as an "economic" QR decomposition.)
| |
| | |
| ==See also==
| |
| * [[Polar decomposition]]
| |
| * [[Eigenvalue decomposition]]
| |
| * [[Spectral decomposition]]
| |
| * [[Matrix decomposition]]
| |
| * [[Zappa–Szép product#Examples|Zappa–Szép product]]
| |
| | |
| ==References==
| |
| <references/>
| |
| * {{Citation | last1=Golub | first1=Gene H. | author1-link=Gene H. Golub | last2=Van Loan | first2=Charles F. | author2-link=Charles F. Van Loan | title=Matrix Computations | publisher=Johns Hopkins | edition=3rd | isbn=978-0-8018-5414-9 | year=1996}}.
| |
| * {{citation | first1=Roger A. | last1=Horn | first2=Charles R. | last2=Johnson | year=1985 | title=Matrix Analysis | publisher=Cambridge University Press | isbn=0-521-38632-2 }}. Section 2.8.
| |
| * {{Citation |last1=Press|first1=WH|last2=Teukolsky|first2=SA|last3=Vetterling|first3=WT|last4=Flannery|first4=BP|year=2007|title=Numerical Recipes: The Art of Scientific Computing|edition=3rd|publisher=Cambridge University Press| publication-place=New York|isbn=978-0-521-88068-8|chapter=Section 2.10. QR Decomposition|chapter-url=http://apps.nrbook.com/empanel/index.html?pg=102}}
| |
| * {{citation | first1=Josef | last1=Stoer | first2=Roland | last2=Bulirsch | year=2002 | title=Introduction to Numerical Analysis | edition=3rd | publisher=Springer | isbn=0-387-95452-X }}.
| |
| | |
| ==External links==
| |
| *[http://www.bluebit.gr/matrix-calculator/ Online Matrix Calculator] Performs QR decomposition of matrices.
| |
| *[http://netlib.org/lapack/lug/node39.html LAPACK users manual] gives details of subroutines to calculate the QR decomposition
| |
| *[http://documents.wolfram.com/mathematica/functions/QRDecomposition Mathematica users manual] gives details and examples of routines to calculate QR decomposition
| |
| *[http://www.alglib.net/ ALGLIB] includes a partial port of the LAPACK to C++, C#, Delphi, etc.
| |
| *[http://eigen.tuxfamily.org/dox-devel/group__QR__Module.html Eigen::QR] Includes C++ implementation of QR decomposition.
| |
| *[http://intopii.com/into/ Into] contains an open source implementation of QR decomposition in C++.
| |
| | |
| {{Numerical linear algebra}}
| |
| | |
| [[Category:Matrix decompositions]]
| |
| [[Category:Numerical linear algebra]]
| |