Agfdhyk

In linear algebra, Cramer's rule is an explicit formula for the solution of a system of linear equations with as many equations as unknowns, valid whenever the system has a unique solution. It expresses the solution in terms of the determinants of the (square) coefficient matrix and of matrices obtained from it by replacing one column by the column vector of right-hand-sides of the equations. It is named after Gabriel Cramer (1704–1752), who published the rule for an arbitrary number of unknowns in 1750,^[1][2] although Colin Maclaurin also published special cases of the rule in 1748^[3] (and possibly knew of it as early as 1729).^[4][5][6]

Cramer's rule implemented in a naïve way is computationally inefficient for systems of more than two or three equations.^[7] In the case of n equations in n unknowns, it requires computation of n + 1 determinants, while Gaussian elimination produces the result with the same computational complexity as the computation of a single determinant.^{[8][9][verification needed]} Cramer's rule can also be numerically unstable even for 2×2 systems.^[10] However, it has recently been shown that Cramer's rule can be implemented in O(n³) time,^[11] which is comparable to more common methods of solving systems of linear equations, such as Gaussian elimination (consistently requiring 2.5 times as many arithmetic operations for all matrix sizes), while exhibiting comparable numeric stability in most cases.

General case

Consider a system of n linear equations for n unknowns, represented in matrix multiplication form as follows:

Ax=b{displaystyle Ax=b}

where the n × n matrix A has a nonzero determinant, and the vector x=(x1,…,xn)T{displaystyle x=(x_{1},ldots ,x_{n})^{mathrm {T} }} is the column vector of the variables. Then the theorem states that in this case the system has a unique solution, whose individual values for the unknowns are given by:

xi=det(Ai)det(A)i=1,…,n{displaystyle x_{i}={frac {det(A_{i})}{det(A)}}qquad i=1,ldots ,n}

where Ai{displaystyle A_{i}} is the matrix formed by replacing the i-th column of A by the column vector b.

A more general version of Cramer's rule^[12] considers the matrix equation

AX=B{displaystyle AX=B}

where the n × n matrix A has a nonzero determinant, and X, B are n × m matrices. Given sequences 1≤i1<i2<…<ik≤n{displaystyle 1leq i_{1}<i_{2}<ldots <i_{k}leq n} and 1≤j1<j2<…<jk≤m{displaystyle 1leq j_{1}<j_{2}<ldots <j_{k}leq m}, let XI,J{displaystyle X_{I,J}} be the k × k submatrix of X with rows in I:=(i1,…,ik){displaystyle I:=(i_{1},ldots ,i_{k})} and columns in J:=(j1,…,jk){displaystyle J:=(j_{1},ldots ,j_{k})}. Let AB(I,J){displaystyle A_{B}(I,J)} be the n × n matrix formed by replacing the is{displaystyle i_{s}} column of A by the js{displaystyle j_{s}} column of B, for all s=1,…,k{displaystyle s=1,ldots ,k}. Then

detXI,J=det(AB(I,J))det(A).{displaystyle det X_{I,J}={frac {det(A_{B}(I,J))}{det(A)}}.}

In the case k=1{displaystyle k=1}, this reduces to the normal Cramer's rule.

The rule holds for systems of equations with coefficients and unknowns in any field, not just in the real numbers.

Proof

The proof for Cramer's rule uses just two properties of determinants: linearity with respect to any given column (taking for that column a linear combination of column vectors produces as determinant the corresponding linear combination of their determinants), and the fact that the determinant is zero whenever two columns are equal (which is implied by the basic property that the sign of the determinant flips if you switch two columns).

Fix the index j of a column. Linearity means that if we consider only column j as variable (fixing the others arbitrarily), the resulting function Rⁿ → R (assuming matrix entries are in R) can be given by a matrix, with one row and n columns, that acts on column j. In fact this is precisely what Laplace expansion does, writing det(A) = C₁a_1,j + ... + C_na_n,j for certain coefficients C₁, ..., C_n that depend on the columns of A other than column j (the precise expression for these cofactors is not important here). The value det(A) is then the result of applying the one-line matrix L_(j) = (C₁C₂ ... C_n) to column j of A. If L_(j) is applied to any other column k of A, then the result is the determinant of the matrix obtained from A by replacing column j by a copy of column k, so the resulting determinant is 0 (the case of two equal columns).

Now consider a system of n linear equations in n unknowns x1,…,xn{displaystyle x_{1},ldots ,x_{n}}, whose coefficient matrix is A, with det(A) assumed to be nonzero:

a11x1+a12x2+⋯+a1nxn=b1a21x1+a22x2+⋯+a2nxn=b2⋮⋮⋮an1x1+an2x2+⋯+annxn=bn.{displaystyle {begin{matrix}a_{11}x_{1}+a_{12}x_{2}+cdots +a_{1n}x_{n}&=&b_{1}\a_{21}x_{1}+a_{22}x_{2}+cdots +a_{2n}x_{n}&=&b_{2}\vdots &vdots &vdots \a_{n1}x_{1}+a_{n2}x_{2}+cdots +a_{nn}x_{n}&=&b_{n}.end{matrix}}}

If one combines these equations by taking C₁ times the first equation, plus C₂ times the second, and so forth until C_n times the last, then the coefficient of x_j will become C₁a_{1, j} + ... + C_na_n,j = det(A), while the coefficients of all other unknowns become 0; the left hand side becomes simply det(A)x_j. The right hand side is C₁b₁ + ... + C_nb_n, which is L_(j) applied to the column vector b of the right hand side b_i. In fact what has been done here is multiply the matrix equation Ax = b on the left by L_(j). Dividing by the nonzero number det(A) one finds the following equation, necessary to satisfy the system:

xj=L(j)⋅bdet(A).{displaystyle x_{j}={frac {L_{(j)}cdot mathbf {b} }{det(A)}}.}

But by construction the numerator is the determinant of the matrix obtained from A by replacing column j by b, so we get the expression of Cramer's rule as a necessary condition for a solution. The same procedure can be repeated for other values of j to find values for the other unknowns.

The only point that remains to prove is that these values for the unknowns, the only possible ones, do indeed together form a solution. But if the matrix A is invertible with inverse A⁻¹, then x = A⁻¹b will be a solution, thus showing its existence. To see that A is invertible when det(A) is nonzero, consider the n × n matrix M obtained by stacking the one-line matrices L_(j) on top of each other for j = 1, ..., n (this gives the adjugate matrix for A). It was shown that L_(j)A = (0 ... 0 det(A) 0 ... 0) where det(A) appears at the position j; from this it follows that MA = det(A)I_n. Therefore,

1det(A)M=A−1,{displaystyle {frac {1}{det(A)}}M=A^{-1},}

completing the proof.

For other proofs, see below.

Finding inverse matrix

Let A be an n × n matrix. Then

Aadj⁡A=(adj⁡A)A=det⁡(A)I{displaystyle A,operatorname {adj} A=(operatorname {adj} A),A=operatorname {det} (A)I}

where adj(A) denotes the adjugate matrix of A, det(A) is the determinant, and I is the identity matrix. If det(A) is invertible in R, then the inverse matrix of A is

A−1=1det⁡(A)adj⁡(A).{displaystyle A^{-1}={frac {1}{operatorname {det} (A)}}operatorname {adj} (A).}

If R is a field (such as the field of real numbers), then this gives a formula for the inverse of A, provided det(A) ≠ 0. In fact, this formula will work whenever R is a commutative ring, provided that det(A) is a unit. If det(A) is not a unit, then A is not invertible.

Applications

Explicit formulas for small systems

Consider the linear system

{a1x+b1y=c1a2x+b2y=c2{displaystyle left{{begin{matrix}a_{1}x+b_{1}y&={color {red}c_{1}}\a_{2}x+b_{2}y&={color {red}c_{2}}end{matrix}}right.}

which in matrix format is

[a1b1a2b2][xy]=[c1c2].{displaystyle {begin{bmatrix}a_{1}&b_{1}\a_{2}&b_{2}end{bmatrix}}{begin{bmatrix}x\yend{bmatrix}}={begin{bmatrix}{color {red}c_{1}}\{color {red}c_{2}}end{bmatrix}}.}

Assume a₁b₂ − b₁a₂ nonzero. Then, with help of determinants, x and y can be found with Cramer's rule as

x=|c1b1c2b2||a1b1a2b2|=c1b2−b1c2a1b2−b1a2,y=|a1c1a2c2||a1b1a2b2|=a1c2−c1a2a1b2−b1a2.{displaystyle {begin{aligned}x&={frac {begin{vmatrix}{color {red}{c_{1}}}&b_{1}\{color {red}{c_{2}}}&b_{2}end{vmatrix}}{begin{vmatrix}a_{1}&b_{1}\a_{2}&b_{2}end{vmatrix}}}={{color {red}c_{1}}b_{2}-b_{1}{color {red}c_{2}} over a_{1}b_{2}-b_{1}a_{2}},quad y={frac {begin{vmatrix}a_{1}&{color {red}{c_{1}}}\a_{2}&{color {red}{c_{2}}}end{vmatrix}}{begin{vmatrix}a_{1}&b_{1}\a_{2}&b_{2}end{vmatrix}}}={a_{1}{color {red}c_{2}}-{color {red}c_{1}}a_{2} over a_{1}b_{2}-b_{1}a_{2}}end{aligned}}.}

The rules for 3 × 3 matrices are similar. Given

{a1x+b1y+c1z=d1a2x+b2y+c2z=d2a3x+b3y+c3z=d3{displaystyle left{{begin{matrix}a_{1}x+b_{1}y+c_{1}z&={color {red}d_{1}}\a_{2}x+b_{2}y+c_{2}z&={color {red}d_{2}}\a_{3}x+b_{3}y+c_{3}z&={color {red}d_{3}}end{matrix}}right.}

which in matrix format is

[a1b1c1a2b2c2a3b3c3][xyz]=[d1d2d3].{displaystyle {begin{bmatrix}a_{1}&b_{1}&c_{1}\a_{2}&b_{2}&c_{2}\a_{3}&b_{3}&c_{3}end{bmatrix}}{begin{bmatrix}x\y\zend{bmatrix}}={begin{bmatrix}{color {red}d_{1}}\{color {red}d_{2}}\{color {red}d_{3}}end{bmatrix}}.}

Then the values of x, y and z can be found as follows:

x=|d1b1c1d2b2c2d3b3c3||a1b1c1a2b2c2a3b3c3|,y=|a1d1c1a2d2c2a3d3c3||a1b1c1a2b2c2a3b3c3|, and z=|a1b1d1a2b2d2a3b3d3||a1b1c1a2b2c2a3b3c3|.{displaystyle x={frac {begin{vmatrix}{color {red}d_{1}}&b_{1}&c_{1}\{color {red}d_{2}}&b_{2}&c_{2}\{color {red}d_{3}}&b_{3}&c_{3}end{vmatrix}}{begin{vmatrix}a_{1}&b_{1}&c_{1}\a_{2}&b_{2}&c_{2}\a_{3}&b_{3}&c_{3}end{vmatrix}}},quad y={frac {begin{vmatrix}a_{1}&{color {red}d_{1}}&c_{1}\a_{2}&{color {red}d_{2}}&c_{2}\a_{3}&{color {red}d_{3}}&c_{3}end{vmatrix}}{begin{vmatrix}a_{1}&b_{1}&c_{1}\a_{2}&b_{2}&c_{2}\a_{3}&b_{3}&c_{3}end{vmatrix}}},{text{ and }}z={frac {begin{vmatrix}a_{1}&b_{1}&{color {red}d_{1}}\a_{2}&b_{2}&{color {red}d_{2}}\a_{3}&b_{3}&{color {red}d_{3}}end{vmatrix}}{begin{vmatrix}a_{1}&b_{1}&c_{1}\a_{2}&b_{2}&c_{2}\a_{3}&b_{3}&c_{3}end{vmatrix}}}.}

Differential geometry

Ricci calculus

Cramer's rule is used in the Ricci calculus in various calculations involving the Christoffel symbols of the first and second kind.^[13]

In particular, Cramer's rule can be used to prove that the divergence operator on a Riemannian manifold is invariant with respect to change of coordinates. We give a direct proof, suppressing the role of the Christoffel symbols.
Let (M,g){displaystyle (M,g)} be a Riemannian manifold equipped with local coordinates (x1,x2,…,xn){displaystyle (x^{1},x^{2},dots ,x^{n})}. Let A=Ai∂∂xi{displaystyle A=A^{i}{frac {partial }{partial x^{i}}}} be a vector field. We use the summation convention throughout.

Theorem.

The divergence of A{displaystyle A},

div⁡A=1detg∂∂xi(Aidetg),{displaystyle operatorname {div} A={frac {1}{sqrt {det g}}}{frac {partial }{partial x^{i}}}left(A^{i}{sqrt {det g}}right),}

is invariant under change of coordinates.

Computing derivatives implicitly

Consider the two equations F(x,y,u,v)=0{displaystyle F(x,y,u,v)=0} and G(x,y,u,v)=0{displaystyle G(x,y,u,v)=0}. When u and v are independent variables, we can define x=X(u,v){displaystyle x=X(u,v)} and y=Y(u,v).{displaystyle y=Y(u,v).}

Finding an equation for ∂x∂u{displaystyle {dfrac {partial x}{partial u}}} is a trivial application of Cramer's rule.

Calculation of ∂x∂u{displaystyle {dfrac {partial x}{partial u}}}

First, calculate the first derivatives of F, G, x, and y: dF=∂F∂xdx+∂F∂ydy+∂F∂udu+∂F∂vdv=0dG=∂G∂xdx+∂G∂ydy+∂G∂udu+∂G∂vdv=0dx=∂X∂udu+∂X∂vdvdy=∂Y∂udu+∂Y∂vdv.{displaystyle {begin{aligned}dF&={frac {partial F}{partial x}}dx+{frac {partial F}{partial y}}dy+{frac {partial F}{partial u}}du+{frac {partial F}{partial v}}dv=0\[6pt]dG&={frac {partial G}{partial x}}dx+{frac {partial G}{partial y}}dy+{frac {partial G}{partial u}}du+{frac {partial G}{partial v}}dv=0\[6pt]dx&={frac {partial X}{partial u}}du+{frac {partial X}{partial v}}dv\[6pt]dy&={frac {partial Y}{partial u}}du+{frac {partial Y}{partial v}}dv.end{aligned}}} Substituting dx, dy into dF and dG, we have: dF=(∂F∂x∂x∂u+∂F∂y∂y∂u+∂F∂u)du+(∂F∂x∂x∂v+∂F∂y∂y∂v+∂F∂v)dv=0dG=(∂G∂x∂x∂u+∂G∂y∂y∂u+∂G∂u)du+(∂G∂x∂x∂v+∂G∂y∂y∂v+∂G∂v)dv=0.{displaystyle {begin{aligned}dF&=left({frac {partial F}{partial x}}{frac {partial x}{partial u}}+{frac {partial F}{partial y}}{frac {partial y}{partial u}}+{frac {partial F}{partial u}}right)du+left({frac {partial F}{partial x}}{frac {partial x}{partial v}}+{frac {partial F}{partial y}}{frac {partial y}{partial v}}+{frac {partial F}{partial v}}right)dv=0\[6pt]dG&=left({frac {partial G}{partial x}}{frac {partial x}{partial u}}+{frac {partial G}{partial y}}{frac {partial y}{partial u}}+{frac {partial G}{partial u}}right)du+left({frac {partial G}{partial x}}{frac {partial x}{partial v}}+{frac {partial G}{partial y}}{frac {partial y}{partial v}}+{frac {partial G}{partial v}}right)dv=0.end{aligned}}} Since u, v are both independent, the coefficients of du, dv must be zero. So we can write out equations for the coefficients: ∂F∂x∂x∂u+∂F∂y∂y∂u=−∂F∂u∂G∂x∂x∂u+∂G∂y∂y∂u=−∂G∂u∂F∂x∂x∂v+∂F∂y∂y∂v=−∂F∂v∂G∂x∂x∂v+∂G∂y∂y∂v=−∂G∂v.{displaystyle {begin{aligned}{frac {partial F}{partial x}}{frac {partial x}{partial u}}+{frac {partial F}{partial y}}{frac {partial y}{partial u}}&=-{frac {partial F}{partial u}}\[6pt]{frac {partial G}{partial x}}{frac {partial x}{partial u}}+{frac {partial G}{partial y}}{frac {partial y}{partial u}}&=-{frac {partial G}{partial u}}\[6pt]{frac {partial F}{partial x}}{frac {partial x}{partial v}}+{frac {partial F}{partial y}}{frac {partial y}{partial v}}&=-{frac {partial F}{partial v}}\[6pt]{frac {partial G}{partial x}}{frac {partial x}{partial v}}+{frac {partial G}{partial y}}{frac {partial y}{partial v}}&=-{frac {partial G}{partial v}}.end{aligned}}} Now, by Cramer's rule, we see that: ∂x∂u=|−∂F∂u∂F∂y−∂G∂u∂G∂y||∂F∂x∂F∂y∂G∂x∂G∂y|.{displaystyle {frac {partial x}{partial u}}={frac {begin{vmatrix}-{frac {partial F}{partial u}}&{frac {partial F}{partial y}}\-{frac {partial G}{partial u}}&{frac {partial G}{partial y}}end{vmatrix}}{begin{vmatrix}{frac {partial F}{partial x}}&{frac {partial F}{partial y}}\{frac {partial G}{partial x}}&{frac {partial G}{partial y}}end{vmatrix}}}.} This is now a formula in terms of two Jacobians: ∂x∂u=−(∂(F,G)∂(u,y))(∂(F,G)∂(x,y)).{displaystyle {frac {partial x}{partial u}}=-{frac {left({frac {partial (F,G)}{partial (u,y)}}right)}{left({frac {partial (F,G)}{partial (x,y)}}right)}}.} Similar formulas can be derived for ∂x∂v,∂y∂u,∂y∂v.{displaystyle {frac {partial x}{partial v}},{frac {partial y}{partial u}},{frac {partial y}{partial v}}.}

Integer programming

Cramer's rule can be used to prove that an integer programming problem whose constraint matrix is totally unimodular and whose right-hand side is integer, has integer basic solutions. This makes the integer program substantially easier to solve.

Ordinary differential equations

Cramer's rule is used to derive the general solution to an inhomogeneous linear differential equation by the method of variation of parameters.

Geometric interpretation

Geometric interpretation of Cramer's rule. The areas of the second and third shaded parallelograms are the same and the second is x1{displaystyle x_{1}} times the first. From this equality Cramer's rule follows.

Cramer's rule has a geometric interpretation that can be considered also a proof or simply giving insight about its geometric nature. These geometric arguments work in general and not only in the case of two equations with two unknowns presented here.

Given the system of equations

a11x1+a12x2=b1a21x1+a22x2=b2{displaystyle {begin{matrix}a_{11}x_{1}+a_{12}x_{2}&=b_{1}\a_{21}x_{1}+a_{22}x_{2}&=b_{2}end{matrix}}}

it can be considered as an equation between vectors

x1(a11a21)+x2(a12a22)=(b1b2).{displaystyle x_{1}{binom {a_{11}}{a_{21}}}+x_{2}{binom {a_{12}}{a_{22}}}={binom {b_{1}}{b_{2}}}.}

The area of the parallelogram determined by (a11a21){displaystyle {binom {a_{11}}{a_{21}}}} and (a12a22){displaystyle {binom {a_{12}}{a_{22}}}} is given by the determinant of the system of equations:

|a11a12a21a22|.{displaystyle {begin{vmatrix}a_{11}&a_{12}\a_{21}&a_{22}end{vmatrix}}.}

In general, when there are more variables and equations, the determinant of n vectors of length n will give the volume of the parallelepiped determined by those vectors in the n-th dimensional Euclidean space.

Therefore, the area of the parallelogram determined by x1(a11a21){displaystyle x_{1}{binom {a_{11}}{a_{21}}}} and (a12a22){displaystyle {binom {a_{12}}{a_{22}}}} has to be x1{displaystyle x_{1}} times the area of the first one since one of the sides has been multiplied by this factor. Now, this last parallelogram, by Cavalieri's principle, has the same area as the parallelogram determined by (b1b2)=x1(a11a21)+x2(a12a22){displaystyle {binom {b_{1}}{b_{2}}}=x_{1}{binom {a_{11}}{a_{21}}}+x_{2}{binom {a_{12}}{a_{22}}}} and (a12a22).{displaystyle {binom {a_{12}}{a_{22}}}.}

Equating the areas of this last and the second parallelogram gives the equation

|b1a12b2a22|=|a11x1a12a21x1a22|=x1|a11a12a21a22|{displaystyle {begin{vmatrix}b_{1}&a_{12}\b_{2}&a_{22}end{vmatrix}}={begin{vmatrix}a_{11}x_{1}&a_{12}\a_{21}x_{1}&a_{22}end{vmatrix}}=x_{1}{begin{vmatrix}a_{11}&a_{12}\a_{21}&a_{22}end{vmatrix}}}

from which Cramer's rule follows.

Other proofs

A proof by abstract linear algebra

This is a restatement of the proof above in abstract language.

Consider the map x→=(x1,…,xn)↦1detA(det(A1),…,det(An)),{displaystyle {vec {x}}=(x_{1},ldots ,x_{n})mapsto {frac {1}{det A}}(det(A_{1}),ldots ,det(A_{n})),} where Ai{displaystyle A_{i}} is the matrix A{displaystyle A} with x→{displaystyle {vec {x}}} substituted in the i{displaystyle i}th column, as in Cramer's rule. Because of linearity of determinant in every column, this map is linear. Observe that it sends the
i{displaystyle i}th column of A{displaystyle A} to the i{displaystyle i}th basis vector e→i=(0,…,1,…,0){displaystyle {vec {e}}_{i}=(0,ldots ,1,ldots ,0)} (with 1 in the i{displaystyle i}th place), because determinant of a matrix with a repeated column is 0. So we have a linear map which agrees with the inverse of A{displaystyle A} on the column space; hence it agrees with A−1{displaystyle A^{-1}} on the span of the column space. Since A{displaystyle A} is invertible, the column vectors span all of Rn{displaystyle mathbb {R} ^{n}}, so our map really is the inverse of A{displaystyle A}. Cramer's rule follows.

A short proof

A short proof of Cramer's rule ^[14] can be given by noticing that x1{displaystyle x_{1}} is the determinant of the matrix

X1=[x100…0x210…0x301…0⋮⋮⋮⋱⋮xn00…1]{displaystyle X_{1}={begin{bmatrix}x_{1}&0&0&dots &0\x_{2}&1&0&dots &0\x_{3}&0&1&dots &0\vdots &vdots &vdots &ddots &vdots \x_{n}&0&0&dots &1end{bmatrix}}}

On the other hand, assuming that our original matrix A is invertible, this matrix X1{displaystyle X_{1}} has columns A−1b,A−1v2,…,A−1vn{displaystyle A^{-1}b,A^{-1}v_{2},ldots ,A^{-1}v_{n}}, where vn{displaystyle v_{n}} is the n-th column of the matrix A. Recall that the matrix A1{displaystyle A_{1}} has columns b,v2,…,vn{displaystyle b,v_{2},ldots ,v_{n}}. Hence we have

x1=det(X1)=det(A−1)det(A1)=det(A1)det(A).{displaystyle x_{1}=det(X_{1})=det(A^{-1})det(A_{1})={frac {det(A_{1})}{det(A)}}.}

The proof for other xj{displaystyle x_{j}} is similar.

Proof using Clifford algebra

Consider the system of three scalar equations in three unknown scalars x1,x2,x3{displaystyle x_{1},x_{2},x_{3}}

a11x1+a12x2+a13x3=c1a21x1+a22x2+a23x3=c2a31x1+a32x2+a33x3=c3{displaystyle {begin{aligned}a_{11}x_{1}+a_{12}x_{2}+a_{13}x_{3}&=c_{1}\a_{21}x_{1}+a_{22}x_{2}+a_{23}x_{3}&=c_{2}\a_{31}x_{1}+a_{32}x_{2}+a_{33}x_{3}&=c_{3}end{aligned}}}

and assign an orthonormal vector basis e1,e2,e3{displaystyle mathbf {e} _{1},mathbf {e} _{2},mathbf {e} _{3}} for G3{displaystyle {mathcal {G}}_{3}} as

a11e1x1+a12e1x2+a13e1x3=c1e1a21e2x1+a22e2x2+a23e2x3=c2e2a31e3x1+a32e3x2+a33e3x3=c3e3{displaystyle {begin{aligned}a_{11}mathbf {e} _{1}x_{1}+a_{12}mathbf {e} _{1}x_{2}+a_{13}mathbf {e} _{1}x_{3}&=c_{1}mathbf {e} _{1}\a_{21}mathbf {e} _{2}x_{1}+a_{22}mathbf {e} _{2}x_{2}+a_{23}mathbf {e} _{2}x_{3}&=c_{2}mathbf {e} _{2}\a_{31}mathbf {e} _{3}x_{1}+a_{32}mathbf {e} _{3}x_{2}+a_{33}mathbf {e} _{3}x_{3}&=c_{3}mathbf {e} _{3}end{aligned}}}

Let the vectors

a1=a11e1+a21e2+a31e3a2=a12e1+a22e2+a32e3a3=a13e1+a23e2+a33e3{displaystyle {begin{aligned}mathbf {a} _{1}&=a_{11}mathbf {e} _{1}+a_{21}mathbf {e} _{2}+a_{31}mathbf {e} _{3}\mathbf {a} _{2}&=a_{12}mathbf {e} _{1}+a_{22}mathbf {e} _{2}+a_{32}mathbf {e} _{3}\mathbf {a} _{3}&=a_{13}mathbf {e} _{1}+a_{23}mathbf {e} _{2}+a_{33}mathbf {e} _{3}end{aligned}}}

Adding the system of equations, it is seen that

c=c1e1+c2e2+c3e3=x1a1+x2a2+x3a3{displaystyle {begin{aligned}mathbf {c} &=c_{1}mathbf {e} _{1}+c_{2}mathbf {e} _{2}+c_{3}mathbf {e} _{3}\&=x_{1}mathbf {a} _{1}+x_{2}mathbf {a} _{2}+x_{3}mathbf {a} _{3}end{aligned}}}

Using the exterior product, each unknown scalar xk{displaystyle x_{k}} can be solved as

c∧a2∧a3=x1a1∧a2∧a3c∧a1∧a3=x2a2∧a1∧a3c∧a1∧a2=x3a3∧a1∧a2x1=c∧a2∧a3a1∧a2∧a3x2=c∧a1∧a3a2∧a1∧a3=a1∧c∧a3a1∧a2∧a3x3=c∧a1∧a2a3∧a1∧a2=a1∧a2∧ca1∧a2∧a3{displaystyle {begin{aligned}mathbf {c} wedge mathbf {a} _{2}wedge mathbf {a} _{3}&=x_{1}mathbf {a} _{1}wedge mathbf {a} _{2}wedge mathbf {a} _{3}\mathbf {c} wedge mathbf {a} _{1}wedge mathbf {a} _{3}&=x_{2}mathbf {a} _{2}wedge mathbf {a} _{1}wedge mathbf {a} _{3}\mathbf {c} wedge mathbf {a} _{1}wedge mathbf {a} _{2}&=x_{3}mathbf {a} _{3}wedge mathbf {a} _{1}wedge mathbf {a} _{2}\x_{1}&={frac {mathbf {c} wedge mathbf {a} _{2}wedge mathbf {a} _{3}}{mathbf {a} _{1}wedge mathbf {a} _{2}wedge mathbf {a} _{3}}}\x_{2}&={frac {mathbf {c} wedge mathbf {a} _{1}wedge mathbf {a} _{3}}{mathbf {a} _{2}wedge mathbf {a} _{1}wedge mathbf {a} _{3}}}={frac {mathbf {a} _{1}wedge mathbf {c} wedge mathbf {a} _{3}}{mathbf {a} _{1}wedge mathbf {a} _{2}wedge mathbf {a} _{3}}}\x_{3}&={frac {mathbf {c} wedge mathbf {a} _{1}wedge mathbf {a} _{2}}{mathbf {a} _{3}wedge mathbf {a} _{1}wedge mathbf {a} _{2}}}={frac {mathbf {a} _{1}wedge mathbf {a} _{2}wedge mathbf {c} }{mathbf {a} _{1}wedge mathbf {a} _{2}wedge mathbf {a} _{3}}}end{aligned}}}

For n equations in n unknowns, the solution for the k-th unknown xk{displaystyle x_{k}} generalizes to

xk=a1∧⋯∧(c)k∧⋯∧ana1∧⋯∧ak∧⋯∧an=(a1∧⋯∧(c)k∧⋯∧an)(a1∧⋯∧ak∧⋯∧an)−1=(a1∧⋯∧(c)k∧⋯∧an)(a1∧⋯∧ak∧⋯∧an)(a1∧⋯∧ak∧⋯∧an)(a1∧⋯∧ak∧⋯∧an)=(a1∧⋯∧(c)k∧⋯∧an)⋅(a1∧⋯∧ak∧⋯∧an)(−1)n(n−1)2(an∧⋯∧ak∧⋯∧a1)⋅(a1∧⋯∧ak∧⋯∧an)=(an∧⋯∧(c)k∧⋯∧a1)⋅(a1∧⋯∧ak∧⋯∧an)(an∧⋯∧ak∧⋯∧a1)⋅(a1∧⋯∧ak∧⋯∧an){displaystyle {begin{aligned}x_{k}&={frac {mathbf {a} _{1}wedge cdots wedge (mathbf {c} )_{k}wedge cdots wedge mathbf {a} _{n}}{mathbf {a} _{1}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{n}}}\&=(mathbf {a} _{1}wedge cdots wedge (mathbf {c} )_{k}wedge cdots wedge mathbf {a} _{n})(mathbf {a} _{1}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{n})^{-1}\&={frac {(mathbf {a} _{1}wedge cdots wedge (mathbf {c} )_{k}wedge cdots wedge mathbf {a} _{n})(mathbf {a} _{1}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{n})}{(mathbf {a} _{1}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{n})(mathbf {a} _{1}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{n})}}\&={frac {(mathbf {a} _{1}wedge cdots wedge (mathbf {c} )_{k}wedge cdots wedge mathbf {a} _{n})cdot (mathbf {a} _{1}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{n})}{(-1)^{frac {n(n-1)}{2}}(mathbf {a} _{n}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{1})cdot (mathbf {a} _{1}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{n})}}\&={frac {(mathbf {a} _{n}wedge cdots wedge (mathbf {c} )_{k}wedge cdots wedge mathbf {a} _{1})cdot (mathbf {a} _{1}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{n})}{(mathbf {a} _{n}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{1})cdot (mathbf {a} _{1}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{n})}}end{aligned}}}

If a_k are linearly independent, then the xk{displaystyle x_{k}} can be expressed in determinant form identical to Cramer’s Rule as

xk=(an∧⋯∧(c)k∧⋯∧a1)⋅(a1∧⋯∧ak∧⋯∧an)(an∧⋯∧ak∧⋯∧a1)⋅(a1∧⋯∧ak∧⋯∧an)=|a1⋅a1⋯a1⋅(c)k⋯a1⋅an⋮⋱⋮⋱⋮ak⋅a1⋯ak⋅(c)k⋯ak⋅an⋮⋱⋮⋱⋮an⋅a1⋯an⋅(c)k⋯an⋅an||a1⋅a1⋯a1⋅ak⋯a1⋅an⋮⋱⋮⋱⋮ak⋅a1⋯ak⋅ak⋯ak⋅an⋮⋱⋮⋱⋮an⋅a1⋯an⋅ak⋯an⋅an|−1=|a1⋮ak⋮an||a1⋯(c)k⋯an||a1⋮ak⋮an|−1|a1⋯ak⋯an|−1=|a1⋯(c)k⋯an||a1⋯ak⋯an|−1=|a11…c1⋯a1n⋮⋱⋮⋱⋮ak1⋯ck⋯akn⋮⋱⋮⋱⋮an1⋯cn⋯ann||a11…a1k⋯a1n⋮|−1{displaystyle {begin{aligned}x_{k}&={frac {(mathbf {a} _{n}wedge cdots wedge (mathbf {c} )_{k}wedge cdots wedge mathbf {a} _{1})cdot (mathbf {a} _{1}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{n})}{(mathbf {a} _{n}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{1})cdot (mathbf {a} _{1}wedge cdots wedge mathbf {a} _{k}wedge cdots wedge mathbf {a} _{n})}}\[8pt]&={begin{vmatrix}mathbf {a} _{1}cdot mathbf {a} _{1}&cdots &mathbf {a} _{1}cdot (mathbf {c} )_{k}&cdots &mathbf {a} _{1}cdot mathbf {a} _{n}\vdots &ddots &vdots &ddots &vdots \mathbf {a} _{k}cdot mathbf {a} _{1}&cdots &mathbf {a} _{k}cdot (mathbf {c} )_{k}&cdots &mathbf {a} _{k}cdot mathbf {a} _{n}\vdots &ddots &vdots &ddots &vdots \mathbf {a} _{n}cdot mathbf {a} _{1}&cdots &mathbf {a} _{n}cdot (mathbf {c} )_{k}&cdots &mathbf {a} _{n}cdot mathbf {a} _{n}end{vmatrix}}{begin{vmatrix}mathbf {a} _{1}cdot mathbf {a} _{1}&cdots &mathbf {a} _{1}cdot mathbf {a} _{k}&cdots &mathbf {a} _{1}cdot mathbf {a} _{n}\vdots &ddots &vdots &ddots &vdots \mathbf {a} _{k}cdot mathbf {a} _{1}&cdots &mathbf {a} _{k}cdot mathbf {a} _{k}&cdots &mathbf {a} _{k}cdot mathbf {a} _{n}\vdots &ddots &vdots &ddots &vdots \mathbf {a} _{n}cdot mathbf {a} _{1}&cdots &mathbf {a} _{n}cdot mathbf {a} _{k}&cdots &mathbf {a} _{n}cdot mathbf {a} _{n}end{vmatrix}}^{-1}\[8pt]&={begin{vmatrix}mathbf {a} _{1}\vdots \mathbf {a} _{k}\vdots \mathbf {a} _{n}end{vmatrix}}{begin{vmatrix}mathbf {a} _{1}&cdots &(mathbf {c} )_{k}&cdots &mathbf {a} _{n}end{vmatrix}}{begin{vmatrix}mathbf {a} _{1}\vdots \mathbf {a} _{k}\vdots \mathbf {a} _{n}end{vmatrix}}^{-1}{begin{vmatrix}mathbf {a} _{1}&cdots &mathbf {a} _{k}&cdots &mathbf {a} _{n}end{vmatrix}}^{-1}\[8pt]&={begin{vmatrix}mathbf {a} _{1}&cdots &(mathbf {c} )_{k}&cdots &mathbf {a} _{n}end{vmatrix}}{begin{vmatrix}mathbf {a} _{1}&cdots &mathbf {a} _{k}&cdots &mathbf {a} _{n}end{vmatrix}}^{-1}\[8pt]&={begin{vmatrix}a_{11}&ldots &c_{1}&cdots &a_{1n}\vdots &ddots &vdots &ddots &vdots \a_{k1}&cdots &c_{k}&cdots &a_{kn}\vdots &ddots &vdots &ddots &vdots \a_{n1}&cdots &c_{n}&cdots &a_{nn}end{vmatrix}}{begin{vmatrix}a_{11}&ldots &a_{1k}&cdots &a_{1n}\vdots end{vmatrix}}^{-1}end{aligned}}}

where (c)_k denotes the substitution of vector a_k with vector c in the k-th numerator position.

Incompatible and indeterminate cases

A system of equations is said to be incompatible or inconsistent when there are no solutions and it is called indeterminate when there is more than one solution. For linear equations, an indeterminate system will have infinitely many solutions (if it is over an infinite field), since the solutions can be expressed in terms of one or more parameters that can take arbitrary values.

Cramer's rule applies to the case where the coefficient determinant is nonzero. In the 2×2 case, if the coefficient determinant is zero, then the system is incompatible if the numerator determinants are nonzero, or indeterminate if the numerator determinants are zero.

For 3×3 or higher systems, the only thing one can say when the coefficient determinant equals zero is that if any of the numerator determinants are nonzero, then the system must be incompatible. However, having all determinants zero does not imply that the system is indeterminate. A simple example where all determinants vanish (equal zero) but the system is still incompatible is the 3×3 system x+y+z=1, x+y+z=2, x+y+z=3.

References

External links

Wikibooks has a book on the topic of: Linear Algebra/Cramer's Rule

• Proof of Cramer's Rule


• WebApp descriptively solving systems of linear equations with Cramer's Rule


• Online Calculator of System of linear equations