Normal, self-adjoint and positive definite operators

In summary, we have two questions that involve the properties of self-adjoint and positive definite operators in a finite dimensional inner product space. The first question asks us to show that the operators UT and TU are diagonalizable with only real eigenvalues. The second question deals with normal operators on a finite dimensional complex inner product space that commute with each other. We need to show that UT* = T*U and find an orthonormal basis of eigenvectors for both T and U. So far, we have tried using the definitions and properties of self-adjoint and positive definite operators, as well as the spectral theorem, but have encountered difficulties.
  • #1
steinmasta
3
0
I have two questions:

1. Let V be a finite dimensional inner product space. Show that if U: V--->V is self-adjoint and T: V---->V is positive definite, then UT and TU are diagonalizable operators with only real eigenvalues.

2. Suppose T and U are normal operators on a finite dimensional complex inner product space V such that TU = UT

a) Show that UT* = T*U
b) show that there is an orthonormal basis for V consisting of vectors that are eigenvectors for both T and U
 
Physics news on Phys.org
  • #2
What have you tried? We cannot help you before you have shown your work.
 
  • #3
What do you know about positive, definite operators, and self-adjoint operators? Do you happen to know what field you're working with?
 
  • #4
Regarding 1, I have tried to show that UT is self-adjoint with respect to the inner product <x,y>' = <T(x),y> but I've had trouble unwinding definitions.

Regarding 2a, I tried to show that if U commutes with T, then U commutes with T*. This corollary to the spectral theorem is useful: If F = C, T is normal iff T* = g(T) for some polynomial g, but I can't seem to make it work.

Regarding 2b, I am not sure where even to begin.
 
  • #5
woah woah woah, don't forget the J in this if you don't know about that then you need a degree and if you've got one...
you need another one
 
  • #6
1. Let V be a finite dimensional inner product space. Show that if U: V--->V is self-adjoint and T: V---->V is positive definite, then UT and TU are diagonalizable operators with only real eigenvalues.

Well you know that U:V--->V is self-adjoint, and therefore has an orthonormal basis of eigenvectors... you know that T:V---->V is Positive, and therefore also self-adjoint, and must also have an orthonormal basis of eigenvectors.

I believe you could also show that UT and TU are Self-adjoint operators from Inner Product algebra.

<Uv,u>=<v,Uu>
<Tv,u>=<v,Tu> and <Tv,v> > 0 for all v in V.
<UTv,u>=<v,(UT)*u>
<UTv,u>=<v,T*U*u>
<UTv,u>=<v,TUu>

(UT)* = T*U*=TU
(TU)*= U*T* = UT

Since U and T are both self-adjoint, TU=UT (Chose the orthonormal basis of eigenvectors, it's the product of the diagonal entries of the respective matrices, and diagonal matrices commute)

I hope that helps. (I might be mistaken, so if anything seems fishy let me know. I'm a freshman and just finished this stuff like three weeks ago... so yeah ;p)
 

FAQ: Normal, self-adjoint and positive definite operators

1. What is the definition of a normal operator?

A normal operator is a linear operator on a complex vector space that commutes with its adjoint, meaning that the operator and its adjoint have the same eigenvectors.

2. How is a self-adjoint operator different from a normal operator?

A self-adjoint operator is a special case of a normal operator, where the operator is equal to its adjoint. This means that all of its eigenvectors are real and the operator is Hermitian.

3. What does it mean for an operator to be positive definite?

A positive definite operator is a self-adjoint operator where all of its eigenvalues are positive. This means that the operator preserves the length of any vector it acts on.

4. What is the significance of normal, self-adjoint, and positive definite operators in linear algebra?

These types of operators have many important properties and applications in linear algebra, including their use in solving differential equations, optimization problems, and quantum mechanics. They also have close connections to the spectral theorem and the diagonalization of matrices.

5. Can a normal operator be both self-adjoint and positive definite?

Yes, a normal operator can be both self-adjoint and positive definite. In fact, all self-adjoint operators are normal, and all positive definite operators are also self-adjoint.

Similar threads

Back
Top