Diagonalizing a Hermitian Matrix: A

  • Thread starter Thread starter pivoxa15
  • Start date Start date
  • Tags Tags
    Hermitian Matrix
Click For Summary

Homework Help Overview

The discussion revolves around finding the diagonal form of a Hermitian matrix using the spectral theorem. The original poster presents a matrix and attempts to apply the theorem, expressing confusion over the resulting diagonal matrix.

Discussion Character

  • Exploratory, Assumption checking, Mathematical reasoning

Approaches and Questions Raised

  • The original poster attempts to diagonalize the matrix using eigenvectors and expresses uncertainty about the correctness of their diagonal matrix. Some participants question the accuracy of the eigenvalues and suggest verifying the calculations.

Discussion Status

Participants are actively discussing the method employed by the original poster, with some providing corrections regarding the eigenvalues and matrix multiplication. There is no explicit consensus, but guidance is being offered regarding the method and potential mistakes.

Contextual Notes

There is mention of the spectral theorem and the relationship between the matrices involved, with some participants questioning the setup of the equation. The original poster's calculations appear to have led to an incorrect diagonal matrix, prompting further exploration of the eigenvalues.

pivoxa15
Messages
2,250
Reaction score
1
Find the diagonal form of the Hermitian matrix

A=\left(<br /> \begin{array}{cc}<br /> 2 &amp; 3i\\<br /> -3i &amp; 2<br /> \end{array}<br /> \right)<br />

The spectral theorem could be used with PAP*=D where D is diagonal matrix and P is a unitary matrix.

I put the columns of P as the eigenvectors (with unit length) of A,

P=\frac{1}{\sqrt{2}}\left(<br /> \begin{array}{cc}<br /> i &amp; -i\\<br /> 1 &amp; 1<br /> \end{array}<br /> \right)<br />

I have checked that P is unitary with P^{-1}=P^{*} and the diagonal entries of D should be 5 and -1. But I got

D=\left(<br /> \begin{array}{cc}<br /> 2 &amp; -3\\<br /> -3 &amp; 2<br /> \end{array}<br /> \right)<br />

which clearly isn't correct.
 
Last edited:
Physics news on Phys.org
Yes, you're right it isn't correct. I'm, nots sure what you want anyone here to do. You have the method correct, so just make sure you'renot making any dumb mistakes in multiplying out matrices.
 
hey
rigth method but wrong eigenvalues

P = 1/sqrt(2) [i -1;i 1]

this will help
 
The spectral theorem could be used with PAP*=D where D is diagonal matrix and P is a unitary matrix.
I can never remember for sure, but isn't it supposed to be A = PDP*? (and thus P*AP = D?)
 
greisen said:
hey
rigth method but wrong eigenvalues

P = 1/sqrt(2) [i -1;i 1]

this will help

That could be my mistake.
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
1K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
6
Views
1K
  • · Replies 2 ·
Replies
2
Views
3K
Replies
9
Views
3K
  • · Replies 1 ·
Replies
1
Views
3K