Math 21b: Linear Algebra Spring 2016
Homework 19: Eigenspaces
This homework is due on Friday, March 25, respectively on Tuesday, March 29, 2016.
1 Find all eigenvalues andeigenvectors for the following matrix A =
2
−1 −1 0 0
−1 2 −1 0 0
−1 −1 4 −1 −1 . What are the algebraic and geometric
0 0 −1 1 0
0 0 −1 0 1
multiplicities? As a hint, we tell you that the eigenvalues 0, 1, 1, 3
appear. Use the trace trick to the 5th eigenvalue.
2 Assume that a 2×2 matrix has trace −5 and determinant 6. Find
its eigenvalues and find a non-diagonal matrix which realizes the
situation.
3 a) Verify that A and AT have the same eigenvalues.
b) Assume A is invertible. What is the relation between the eigen-
values of A and A−1?
4 The vector Anb gives pollution levels in the Silvaplana, Sils and
St Moritz lake n weeks after an oil spill. The matrix is A =
0.7 0 0 100
and b = 0 is the initial pollution level. Find
0.1 0.6 0
0 0.2 0.8 0
a closed form solution for the pollution after n weeks.
5 1) Find all the eigenvalues and eigenvectors of the matrix
1
0 0 0
1 2 0 0
A= .
1 6 6 0
1 14 36 24
b) Find the eigenvectors of A3, where A is the previous matrix.
c) Find the eigenvectors of (AT )−1, where A is the previous ma-
trix.
Eigenspaces
A nonzero vector ~v is called an eigenvector, if Av = λv for
some λ. The set of eigenvectors is called the eigenspace Eλ.
It is the kernel of A − λIn. The dimension of the eigenspace is
called the geometric multiplicity of λ. There is a general
result which tells that the geometric multiplicity of λ is always
smaller or equal to the algebraic multiplicity.
Recall that A is similar to B if there exists an invertible S such
that B = S −1AS. If A and B are similar, then they have the
same characteristic polynomial, the same eigenvalues and alge-
braic multiplicities as well as the same geometric multiplicities.
Similar matrices also have the same trace tr(A) = λ1 + ... + λn
as well as determinant det(A) = λ1 · · · λn. These formulas hold
in general if we allow the eigenvalues λi to be complex. More
on complex eigenvalues next week.