跳到主要内容

Elementary Matrix Operations and Systems of Linear Equations

Elementary matrix operations and elementary matrices

Let AA be an m×nm \times n matrix. Any one of the following three operations on the rows [columns] of AA is called an elementary row [column] operation:

  1. interchanging any two rows [columns] of AA;

  2. multiplying any row [column] of AA by a nonzero scalar;

  3. adding any scalar multiple of a row [column] of AA to another row [column].

An n×nn \times n elementary matrix is a matrix obtained by performing an elementary operation on InI_n. The elementary matrix is said to be of type 1, 2, or 3 according to whether the elementary operation performed on InI_n is a type 1, 2, or 3 operation, respectively.

Theorems

  1. Let AMm×n(F)A \in M_{m \times n}(F), and suppose that BB is obtained from AA by performing an elementary row [column] operation. Then there exists an m×mm \times m [n×nn \times n] elementary matrix EE such that B=EAB = EA [B=AEB = AE]. In fact, EE is obtained from ImI_m [InI_n] by performing the same elementary row [column] operation as that which was performed on AA to obtain BB.

  2. Elementary matrices are invertible, and the inverse of an elementary matrix is an elementary matrix of the same type.

The rank of a matrix and matrix inverses

If AMm×n(F)A \in M_{m \times n}(F), we define the rank of AA, denoted rank(A)\operatorname{rank}(A), to be the rank of the linear transformation LA:FnFmL_{A} : F^{n} \to F^{m}.

Let AA and BB be m×nm \times n and m×pm \times p matrices, respectively. By the augmented matrix (AB)(A \mid B), we mean the m×(n+p)m \times (n + p) matrix (A B)(A\ B), that is, the matrix whose first nn columns are the columns of AA, and whose last pp columns are the columns of BB.

Theorems

  1. Let T:VWT : V \to W be a linear transformation between finite-dimensional vector spaces, and let β\beta and γ\gamma be ordered bases for VV and WW, respectively. Then
rank(T)=rank([T]βγ).\begin{equation*} \operatorname{rank}(T) = \operatorname{rank}([T]_{\beta}^{\gamma}). \end{equation*}
  1. Let AA be an m×nm \times n matrix. If PP and QQ are invertible m×mm \times m and n×nn \times n matrices, respectively, then

    (a) rank(AQ)=rank(A)\operatorname{rank}(AQ) = \operatorname{rank}(A),

    (b) rank(PA)=rank(A)\operatorname{rank}(PA) = \operatorname{rank}(A), and therefore,

    (c) rank(PAQ)=rank(A)\operatorname{rank}(PAQ) = \operatorname{rank}(A).

  2. The rank of any matrix equals the maximum number of its linearly independent columns; that is, the rank of a matrix is the dimension of the subspace generated by its columns.

  3. Let AA be an m×nm \times n matrix of rank rr. Then rmr \le m, rnr \le n, and, by means of a finite number of elementary row and column operations, AA can be transformed into the matrix

    D=(IrO1O2O3),\begin{equation*} D = \begin{pmatrix} I_r & O_1 \\ O_2 & O_3 \end{pmatrix}, \end{equation*}

    where O1O_1, O2O_2, and O3O_3 are zero matrices. Thus Dii=1D_{ii} = 1 for iri \le r and Dij=0D_{ij} = 0 otherwise.

  4. Let AA be an m×nm \times n matrix of rank rr. Then there exist invertible matrices BB and CC of sizes m×mm \times m and n×nn \times n, respectively, such that D=BACD = BAC, where

D=(IrO1O2O3),\begin{equation*} D = \begin{pmatrix} I_r & O_1 \\ O_2 & O_3 \end{pmatrix}, \end{equation*}

​ is the m×nm \times n matrix in which O1O_1, O2O_2, and O3O_3 are zero matrices.

  1. (a) rank(At)=rank(A)\operatorname{rank}(A^{t}) = \operatorname{rank}(A).

    (b) The rank of any matrix equals the maximum number of its linearly independent rows; that is, the rank of a matrix is the dimension of the subspace generated by its rows.

    (c) The rows and columns of any matrix generate subspaces of the same dimension, numerically equal to the rank of the matrix.

  2. (a) rank(UT)rank(U)\operatorname{rank}(UT) \le \operatorname{rank}(U).

    (b) rank(UT)rank(T)\operatorname{rank}(UT) \le \operatorname{rank}(T).

    (c) rank(AB)rank(A)\operatorname{rank}(AB) \le \operatorname{rank}(A).

    (d) rank(AB)rank(B)\operatorname{rank}(AB) \le \operatorname{rank}(B).

Systems of linear equations

a11x1+a12x2++a1nxn=b1a21x1+a22x2++a2nxn=b2 am1x1+am2x2++amnxn=bm,\begin{equation*} \begin{aligned} 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 \\ a_{m1}x_{1} + a_{m2}x_{2} + \cdots + a_{mn}x_{n} &= b_{m}, \end{aligned} \end{equation*}

where aija_{ij} and bib_{i} (1im1 \le i \le m and 1jn1 \le j \le n) are scalars in a field FF and x1,x2,,xnx_{1}, x_{2}, \ldots, x_{n} are nn variables taking values in FF, is called a system of mm linear equations in nn unknowns over the field FF.

The m×nm \times n matrix

A=(a11a12a1na21a22a2nam1am2amn)\begin{equation*} A = \begin{pmatrix} a_{11} & a_{12} & \cdots & a_{1n} \\ a_{21} & a_{22} & \cdots & a_{2n} \\ \vdots & \vdots & \ddots & \vdots \\ a_{m1} & a_{m2} & \cdots & a_{mn} \end{pmatrix} \end{equation*}

is called the coefficient matrix of the system (S)(S).

If we let

x=(x1x2xn)andb=(b1b2bm),\begin{equation*} x = \begin{pmatrix} x_{1} \\ x_{2} \\ \vdots \\ x_{n} \end{pmatrix} \qquad\text{and}\qquad b = \begin{pmatrix} b_{1} \\ b_{2} \\ \vdots \\ b_{m} \end{pmatrix}, \end{equation*}

then the system (S)(S) may be rewritten as a single matrix equation

Ax=b.\begin{equation*} Ax = b. \end{equation*}

To exploit the results that we have developed, we often consider a system of linear equations as a single matrix equation. A solution to the system (S)(S) is an nn-tuple

s=(s1s2sn)Fn\begin{equation*} s = \begin{pmatrix} s_{1} \\ s_{2} \\ \vdots \\ s_{n} \end{pmatrix} \in F^{n} \end{equation*}

such that As=bAs = b. The set of all solutions to the system (S)(S) is called the solution set of the system. System (S)(S) is called consistent if its solution set is nonempty; otherwise it is called inconsistent.

A system Ax=bAx = b of mm linear equations in nn unknowns is said to be homogeneous if b=0b = 0. Otherwise the system is said to be nonhomogeneous.

Any homogeneous system has at least one solution, namely, the zero vector.

Theorems

  1. Let KK be the solution set of a system of linear equations Ax=bAx = b, and let KHK_{H} be the solution set of the corresponding homogeneous system Ax=0Ax = 0. Then for any solution ss to Ax=bAx = b, K={s}+KH={s+k:kKH}.\begin{equation*} K = \{s\} + K_{H} = \{\, s + k : k \in K_{H} \,\}. \end{equation*}

  1. Let Ax=0Ax = 0 be a homogeneous system of mm linear equations in nn unknowns over a field FF. Let KK denote the set of all solutions to Ax=0Ax = 0. Then

    K=N(LA),\begin{equation*} K = N(L_A), \end{equation*}

    hence KK is a subspace of FnF^n of dimension

    nrank(LA)=nrank(A).\begin{equation*} n - \mathrm{rank}(L_A) = n - \mathrm{rank}(A). \end{equation*}
  2. Let Ax=bAx = b be a system of nn linear equations in nn unknowns. If AA is invertible, then the system has exactly one solution, namely A1bA^{-1}b. Conversely, if the system has exactly one solution, then AA is invertible.

  3. Let Ax=bAx = b be a system of linear equations. Then the system is consistent if and only if

    rank(A)=rank(Ab).\begin{equation*} \mathrm{rank}(A) = \mathrm{rank}(A|b). \end{equation*}

Gaussian elimination

A matrix is said to be in reduced row echelon form if the following three conditions are satisfied:

(a) Any row containing a nonzero entry precedes any row in which all the entries are zero (if any).

(b) The first nonzero entry in each row is the only nonzero entry in its column.

(c) The first nonzero entry in each row is 11 and it occurs in a column to the right of the first nonzero entry in the preceding row.

The reduced row echelon form of a matrix is unique.

Let Ax=bAx = b be a system of rr nonzero equations in nn unknowns. Suppose that rank(A)=rank(Ab)\operatorname{rank}(A) = \operatorname{rank}(A|b) and that (Ab)(A|b) is in reduced row echelon form. Then

(a) rank(A)=r\operatorname{rank}(A) = r.

(b) If the general solution obtained by the procedure above is of the form

s=s0+t1u1+t2u2++tnrunr,\begin{equation*} s = s_0 + t_1 u_1 + t_2 u_2 + \cdots + t_{\,n-r}\,u_{\,n-r}, \end{equation*}

then {u1,u2,,unr}\{u_1, u_2, \ldots, u_{\,n-r}\} is a basis for the solution set of the corresponding homogeneous system, and s0s_0 is a solution to the original system.

LU decomposition

Let AA be an m×nm \times n matrix. Suppose that AA admits an LU decomposition of the form

A=LU,\begin{equation*} A = LU, \end{equation*}

where

  • LL is an m×mm \times m lower triangular matrix with 11's on the diagonal, and
  • UU is an m×nm \times n upper triangular matrix.

We consider the system

Ax=b.\begin{equation*} Ax = b. \end{equation*}

Because A=LUA = LU, the system becomes

LUx=b.\begin{equation*} LUx = b. \end{equation*}

Let

y=Ux.\begin{equation*} y = Ux. \end{equation*}

Then the solution process consists of the following two steps.