24小时热门版块排行榜    

查看: 1491  |  回复: 21

zaq123321

专家顾问 (著名写手)

Assume A and B are real square matrices.

[ 发自手机版 http://muchong.com/3g ]
小木虫给我温暖,给我希望,爱就要爱小木虫。
11楼2014-12-13 20:02:16
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

zaq123321

专家顾问 (著名写手)

★ ★
feixiaolin: 金币+2 2014-12-14 07:54:46
引用回帖:
2楼: Originally posted by hank612 at 2014-12-11 06:28:31
1. 任取可逆矩阵A,  那么I 的特征值和 A及A^{-1} 的特征值  毫无关系.

2. I +DM 的特征值 当然是 1+(DM特征值)

可是由于所有的矩阵A都在可以找到一个酉矩阵U, 使的 U^T * A * U = D * M

其中D 是对角阵,  ...

Can you give the proof of

所有的矩阵A都在可以找到一个酉矩阵U, 使的 U^T * A * U = D * M

其中D 是对角阵, M 是对称阵

[ 发自手机版 http://muchong.com/3g ]
小木虫给我温暖,给我希望,爱就要爱小木虫。
12楼2014-12-13 20:49:49
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

hank612

至尊木虫 (著名写手)

★ ★
feixiaolin: 金币+2 2014-12-14 07:54:54
引用回帖:
12楼: Originally posted by zaq123321 at 2014-12-13 20:49:49
Can you give the proof of

所有的矩阵A都在可以找到一个酉矩阵U, 使的 U^T * A * U = D * M

其中D 是对角阵, M 是对称阵
...

我不知道具体证明'  Wikipedia 上 symmetric matrix 中有一节提到该命题'
还有索引'引文中说这是Frobenius的结果'应该挺靠谱的
We_must_know. We_will_know.
13楼2014-12-14 04:55:39
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

zaq123321

专家顾问 (著名写手)

引用回帖:
13楼: Originally posted by hank612 at 2014-12-14 04:55:39
我不知道具体证明'  Wikipedia 上 symmetric matrix 中有一节提到该命题'
还有索引'引文中说这是Frobenius的结果'应该挺靠谱的...

Can you post the webpage? I searched wikipedia but didn't find it.

http://en.wikipedia.org/wiki/Symmetric_matrix
小木虫给我温暖,给我希望,爱就要爱小木虫。
14楼2014-12-14 06:58:47
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

hank612

至尊木虫 (著名写手)

引用回帖:
14楼: Originally posted by zaq123321 at 2014-12-14 06:58:47
Can you post the webpage? I searched wikipedia but didn't find it.

http://en.wikipedia.org/wiki/Symmetric_matrix...

"   Decomposition[edit]
Using the Jordan normal form, one can prove that every square real matrix can be written as a product of two real symmetric matrices, and every square complex matrix can be written as a product of two complex symmetric matrices.[4]
"
in the link you given: http://en.wikipedia.org/wiki/Symmetric_matrix
请教几个关于特征值的问题。
symmetric matrix.png

We_must_know. We_will_know.
15楼2014-12-14 08:14:49
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

全全fly

新虫 (小有名气)

引用回帖:
5楼: Originally posted by hank612 at 2014-12-13 03:33:59
考虑A=diag(a, a^{-1}), 2x2矩阵B=(a^{-1}+c, c; a-a^{-1}-c, a-c)

那么A 和B 的特征值都是 a, a^{-1}.

若c=\frac{x+x^{-1}-2}{a-a^{-1}},  那么AB 的特征值是 x 与 x^{-1}.

由x 的任意性, 任何想从a 推 ...

谢谢
我啥都不懂
16楼2014-12-14 09:17:13
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

全全fly

新虫 (小有名气)

引用回帖:
7楼: Originally posted by pippi6 at 2014-12-13 09:30:57
嗯,非常有意思的例子。我也验证了一下,确实, AB 的特征值不必是A的特征值乘以B的特征值。...

事实上,条件加强一下,这个关系是成立的:
若 A B是半正定的Hermite 阵,则
lamda_n(A)lamda_i(B)<=lamda_i(AB)<=lamda_1(A)lamda_i(B)
或者
lamda_i(A)lamda_n(B)<=lamda_i(AB)<=lamda_i(A)lamda_1(B)
(lamda_i(A)的表示A的特征值,lamda_1(A)>=........lamda_n(A))
我啥都不懂
17楼2014-12-14 09:30:48
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

zaq123321

专家顾问 (著名写手)

引用回帖:
15楼: Originally posted by hank612 at 2014-12-14 08:14:49
"   Decomposition
Using the Jordan normal form, one can prove that every square real matrix can be written as a product of two real symmetric matrices, and every square complex matrix can be w ...

It said any matrix can be written as the product of two symmetric matrices. But your claim is different with that one. How to go further to get your claim? Thanks.
小木虫给我温暖,给我希望,爱就要爱小木虫。
18楼2014-12-14 09:52:10
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

zaq123321

专家顾问 (著名写手)

【答案】应助回帖

1. On spectral radius, we have r(AB)<= r(A)r(B).

[ 发自手机版 http://muchong.com/3g ]
小木虫给我温暖,给我希望,爱就要爱小木虫。
19楼2014-12-14 23:14:56
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

pippi6

铁杆木虫 (著名写手)

工程和科学数值计算咨询

引用回帖:
10楼: Originally posted by zaq123321 at 2014-12-13 19:59:02
Let's do some math together. The product  of eigenvalu was of AB is x*x^(-1)=1. The product of eigenvalues of A and eigenvalues of B is a*a^(-1)*a*a^(-1)=1. So 1=1. I think you misunderstood my clai ...

My appology I overlooked that you indeed said "products of eigenvalues of AB". If you were talking about products of eigenvalues of AB, that is simply the determinant of AB. So your claim was simply that determinant of AB equals determinant of A multiplied by  determinant of B, and don't you think that is a little too trivial? The point of the discussion here is the collection of single eigenvalues of AB, and their relation to eigenvalues of A and B.
20楼2014-12-15 08:58:09
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖
相关版块跳转 我要订阅楼主 全全fly 的主题更新
信息提示
请填处理意见