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.

Going through this paper:

http://dx.doi.org/10.1016/S0893-9659(00)00169-5

at the bottom of page 407, the authors seem to compute the determinant of a matrix by expanding down the diagonal.

The authors discuss a matrix

$A = \left[ \begin{array}{ccc} a_{11} a_{12} a_{13} \\ a_{21} a_{22} a_{23} \\ a_{31} a_{32} a_{33} \end{array}\right]$

and call $M_{ij}$ the cofactor of $a_{ij}$.

Then at the bottom of page 407 they write

$\det(A) = a_{11} M_{11} + a_{22} M_{22} + a_{33} M_{33}.$

Is expansion down the diagonal possible? Is it possible in some special cases?

share|improve this question
2  
This type of expansion is certainly not possible for all matrices. Consider $2\times 2$ matrices for instance... –  Abel Jun 11 '13 at 14:11
1  
The equation on page 407 has some $d$'s in it, which I don't understand. They might represent differentials rather than factors, but the equation doesn't look correct under that interpretation either. –  Andreas Blass Jun 11 '13 at 14:53
    
@AndreasBlass I thought the d's were diffusion coefficients but you may be correct -- however, as you say, this would indicate that det(A) is somehow a function of the diagonal entries of A and that doesn't seem correct. –  Name Jun 11 '13 at 15:13

1 Answer 1

up vote 3 down vote accepted

You are not quoting the formula correctly. It says $$ \color{red}{d}[\det(A)] = M_{11}\color{red}{d}a_{11} + M_{22}\color{red}{d}a_{22} + M_{33}\color{red}{d}a_{33},\tag{1} $$ where the $\color{red}{d}$ means derivative. The authors are being sloppy here. They actually consider a matrix function $A(t)=A(0)-tD$, where $A(0)$ is a constant and stable matrix, $D$ is a constant positive diagonal matrix and $t=k^2\ge0$. The $A$ in $(1)$ is actually $A(t)$ and the $d$ means derivative with respect to $t$. Formula $(1)$ is a direct consequence of Jacobi's formula.

share|improve this answer

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.