Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

If I have 2 congruent $n\times n$ matrices $A,B$ both diagonalized with all non-zero eigenvalues and the eigenvalues are ordered in descending order down the diagonal. $a$ is the number of positive eigenvalues of $A$ and $b$ that of $B$. What then is the must the $a=b$? I know that there is a theorem that says it is so, but I am not allowed to use that. I am supposed to argue by congruence.

Thank you.

share|improve this question
    
"I am supposed to argue by congruence".....Homework? Please tag it so. In a nutshell, are you asking whether 2 congruent non-singular diagonal matrices have the same number of positive eigenvalues? –  Inquest May 23 '12 at 10:39
    
@Nunoxic: Tagged. :) Yes. –  greg May 23 '12 at 11:53
add comment

1 Answer

up vote 1 down vote accepted

I will show this for the $2 \times 2$ case by explicitly doing the matrix multiplication. You can either generalise this to the $n \times n$ case yourself, or wait for someone with a more elegant answer.

Matrices $A$ and $B$ are congruent, i.e. there exists an invertible matrix $P$ such that: $$ P^{T}AP = B.$$ Write out the elements of the matrices and perform the matrix multiplication, i.e. $$ \left( \begin{array}{cc} p_{11} & p_{21} \\ p_{12} & p_{22} \end{array}\right)\left( \begin{array}{cc} a_{11} & 0 \\ 0 & a_{22} \end{array}\right)\left( \begin{array}{cc} p_{11} & p_{12} \\ p_{21} & p_{22} \end{array}\right) = \left( \begin{array}{cc} b_{11} & 0 \\ 0 & b_{22} \end{array}\right),$$ $$\left( \begin{array}{cc} p_{11}^{2}a_{11} + p_{21}^{2}a_{22} & p_{11}p_{12}a_{11} + p_{21}p_{22}a_{22} \\ p_{11}p_{12}a_{11} + p_{21}p_{22}a_{22} & p_{12}^{2}a_{11} + p_{22}^{2}a_{22}\end{array}\right) = \left( \begin{array}{cc} b_{11} & 0 \\ 0 & b_{22} \end{array}\right).$$ To satisfy this equation let $p_{12} = p_{21}= 0$. In this case $b_{jj} = a_{jj}p_{jj}^{2}$, i.e. the signs of the $j$-th entries in $A$ and $B$ are the same.

share|improve this answer
    
I like this way and I believe it can be extended to a general n x n case. +1. –  Inquest May 23 '12 at 10:44
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.