2
$\begingroup$

Let $A,B\in M_n(\mathbb{R})$ and $U=A+iB$ unitary. $R=diag(r_1,r_2,…,r_n)$ is a diagonal matrix with $r_i>0, \forall i $. I need to calculate $\det(Ae^{-R}A^T+Be^{R}B^T)$. This matrix $Ae^{-R}A^T+Be^{R}B^T$ can be written as $XX^*$ where $X=Ae^{-R}+iBe^{R}$ and $X^*$ denotes the conjugate transpose of X. Is there a way to express the eigenvalues of $X$ in terms of eigenvalues of $U$?

(Is there a better way of calculating $\det(Im(\gamma Z))$ where $\gamma=\big(\begin{smallmatrix}A&B\\-B&A\end{smallmatrix}\big)\in Sp_{2n}(\mathbb{R})\bigcap SO_{2n}(\mathbb{R})$ ,and $Z=\big(\begin{smallmatrix}e^R&0\\0&e^{-R}\end{smallmatrix}\big)\cdot i \in \mathbb{H}_n$, the Siegel Upper Half Plane of degree $n$.)

Improve this question
$\endgroup$
1
  • 1
    $\begingroup$ It is a bit difficult for me to understand the genuine question here. There is the old identity $\det \Im \gamma z=|\det (cz+d)|^{-2}\cdot \det \Im z$ which applies as well in the Siegel modular case as in the elliptic modular case. If the question is about better algorithms, I have no idea. $\endgroup$
    – paul garrett
    Apr 27, 2016 at 23:24

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.