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 $A, B$ are positive definite matrices of order $n$, what is the smallest constant $c$ such that $\det(A^2+B^2)\le c\det(A+B)^2$?

share|improve this question
    
If such $c$ does not exist, it must depend on the size $n$. –  Sunni Jun 1 '12 at 20:16
1  
Have you tried the case where $A$ and $B$ are simultaneously diagonalizable? Might be an easy first step. –  Rahul Jun 1 '12 at 20:24
    
@RahulNarain: When AB=BA, we may take $c=1$. I guess, generally, $c>1$. –  Sunni Jun 1 '12 at 20:57

1 Answer 1

up vote 1 down vote accepted

There is no such $c$. Consider the $2 \times 2$ positive definite matrices (for $t \ne 0$) $$ A = \pmatrix{1 & 0\cr 0 & t^2\cr}, \ B = \pmatrix{1 & t\cr t & 2 t^2\cr}$$ Then $$\dfrac{\det(A^2 + B^2)}{\det(A+B)^2} = {\dfrac {1+7\,{t}^{2}+{t}^{4}}{25 {t}^{2}}} \to \infty \ \text{as}\ t \to 0$$

share|improve this answer
    
Isn't the determinant of $A+B$ in this case $5t^2$, and so the denominator should actually be $25t^4$, giving a constant ratio as $t\rightarrow\infty$? –  mjqxxxx Jun 2 '12 at 0:04
    
The answer is correct. –  Sunni Jun 2 '12 at 1:05
    
Yes $\det(A+B) = 5 t^2$, but $\det(A^2+B^2)$ also has a factor of $t^2$. –  Robert Israel Jun 2 '12 at 1:22

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.