A
\mathbf A
A有两行相同,则必有
d
e
t
(
A
)
=
0
det(\mathbf A)=0
det(A)=0 证明:交换相同的两行,行列式的样子不变值也应不变,则通过性质2,
d
e
t
(
A
)
=
−
d
e
t
(
A
)
⇒
d
e
t
(
A
)
=
0
det(\mathbf A)=-det(\mathbf A)\Rightarrow det(\mathbf A)=0
det(A)=−det(A)⇒det(A)=0
矩阵
A
\mathbf A
A不可逆/为奇异矩阵
⟺
\iff
⟺
d
e
t
(
A
)
=
0
det(\mathbf A)=0
det(A)=0 证明:,那么矩阵
A
\mathbf A
A不可逆一切行列式可以“消元”为上三角阵,若有“主元”为0,这对应了不可逆,也对应了
d
e
t
(
A
)
=
0
det(\mathbf A)=0
det(A)=0
小结:结合3和5,只要有全零行(存在非零列向量
x
\boldsymbol x
x使
A
x
=
0
\mathbf A \boldsymbol x=\boldsymbol 0
Ax=0),一定不可逆,必有
d
e
t
(
A
)
=
0
det(\mathbf A)=0
det(A)=0 即使没有全零行,也可推广:只要矩阵列向量线性相关,同样不可逆,有
d
e
t
(
A
)
=
0
det(\mathbf A)=0
det(A)=0
d
e
t
(
A
B
)
=
d
e
t
(
A
)
d
e
t
(
B
)
det(\mathbf A\mathbf B)=det(\mathbf A)det(\mathbf B)
det(AB)=det(A)det(B) 推论:
d
e
t
(
A
−
1
)
=
d
e
t
(
I
)
/
d
e
t
(
A
)
=
1
/
d
e
t
(
A
)
det(\mathbf A^{-1})=det(\mathbf I)/det(\mathbf A)=1/det(\mathbf A)
det(A−1)=det(I)/det(A)=1/det(A),注意式子仅当
A
\mathbf A
A可逆时/
d
e
t
(
A
)
≠
0
det(\mathbf A)\neq 0
det(A)=0时成立
d
e
t
(
A
2
)
=
(
d
e
t
(
A
)
)
2
det(\mathbf A^{2})=(det(\mathbf A))^2
det(A2)=(det(A))2,但注意
d
e
t
(
2
A
)
=
2
n
d
e
t
(
A
)
det(2\mathbf A)=2^ndet(\mathbf A)
det(2A)=2ndet(A)(原始性质3①,每行提出倍数)
d
e
t
(
A
T
)
=
d
e
t
(
A
)
det(\mathbf A^{T})=det(\mathbf A)
det(AT)=det(A) 证明:矩阵消元对应LU分解
A
=
L
U
\mathbf{A=LU}
A=LU,则就是要证明
d
e
t
(
U
T
L
T
)
=
d
e
t
(
L
U
)
det(\mathbf U^T \mathbf L^T)=det(\mathbf L \mathbf U)
det(UTLT)=det(LU),就是证明
d
e
t
(
U
T
)
d
e
t
(
L
T
)
=
d
e
t
(
L
)
d
e
t
(
U
)
det(\mathbf U^T)det(\mathbf L^T)=det(\mathbf L)det(\mathbf U)
det(UT)det(LT)=det(L)det(U),由于分解后的两个矩阵
L
\mathbf L
L和
U
\mathbf U
U都是三角矩阵,根据上述性质4显然有
d
e
t
(
U
T
)
=
d
e
t
(
U
)
det(\mathbf U^T)=det(\mathbf U)
det(UT)=det(U)、
d
e
t
(
L
T
)
=
d
e
t
(
L
)
det(\mathbf L^T)=det(\mathbf L)
det(LT)=det(L) 推论:上述所有关于“行”的性质,通过转置可以得到对应的“列”的性质,这里不再一一列举