Check the spectral theorem for this matrix

  • Thread starter Thread starter LCSphysicist
  • Start date Start date
  • Tags Tags
    Matrix Theorem
LCSphysicist
Messages
644
Reaction score
162
Homework Statement
.
Relevant Equations
.
1615235843259.png

I found three projection operators
$$P_{1}=
\begin{pmatrix}
1/2 & & \\
& -\sqrt{2}/2 & \\
& & 1/2
\end{pmatrix}$$
$$P_{2}=
\begin{pmatrix}
1/2 & & \\
& \sqrt{2}/2 & \\
& & 1/2
\end{pmatrix}$$
$$P_{3}=
\begin{pmatrix}
-1/\sqrt{2} & & \\
& & \\
& & 1/\sqrt{2}
\end{pmatrix}$$

From this five properties
1615235980298.png
, i am having trouble to prove the ii, iii, and iv. I mean, i could understand this conditions in the continuous situation, in which we can make the "eigenvalue tends to" something. But how to interpret it when we are dealing with discrete conditions? In which we have really just three eigenvalues.
 

Attachments

  • 1615231525274.png
    1615231525274.png
    6.6 KB · Views: 176
  • 1615235888901.png
    1615235888901.png
    867 bytes · Views: 157
Physics news on Phys.org
What is the definition of E in those statements?
 
##E(\lambda)## is an projection (operator) valued function. In the case of discrete eigenvalues it is a step function which has a step at each eigenvalue . ##dE(\lambda)/d\lambda## has a ##\delta## function spike at each eigenvalue, which gives you the discrete eigenvalue when you integrate.
 
There are two things I don't understand about this problem. First, when finding the nth root of a number, there should in theory be n solutions. However, the formula produces n+1 roots. Here is how. The first root is simply ##\left(r\right)^{\left(\frac{1}{n}\right)}##. Then you multiply this first root by n additional expressions given by the formula, as you go through k=0,1,...n-1. So you end up with n+1 roots, which cannot be correct. Let me illustrate what I mean. For this...

Similar threads

Back
Top