Based on the selection of questions set in past years, we should be able to do each of these problems set out below. Assume a non-singular matrix A has an eigenvector e and associated eigenvalue v. Let D be a diagonal matrix. We have shown that A = PDP-1 where P is constructed from the values of e and that the diagonal of D includes the various values of v.
1 Show A10= PD10 P-1. Comment on Aⁿ. If necessary use the matrix A below.
2 State the eigenvectors and corresponding eigenvalues of kA, for real number k.
3 Let A be the 3x3 matrix with rows (5 4 1) (-6 -2 3) (8 8 3) and Q be the matrix with rows (1 -2 1) (2 -1 -1) (-1 1 2), possibly as shown below:
5 4 1 1 -2 1
A is -6 -2 3 Q is 2 -1 -1
8 8 3 -1 1 2
a) Convert A into a cubic equation.
b) Given A = PDP-1, find
c) Matrix B is such that B = QAQ-1. Find the eigenvalues and eigenvectors for B, where Q is shown above.
4 Given matrix A has eigenvectors ei and eigenvalues vi, and B has ei and wi, then
a. Show A+B has ei and vi + wi
b. Find eigenvectors and eigenvalues for aA + bB
c. Find eigenvectors and eigenvalues for (aA + bB)⁴
d. What about C = A + bI ?
e. What about A + B² ? A + C² ?
5 Consider the inverse matrix A-1, where A, e and v are defined as before
a. Is e an eigenvector for A-1?
b. Is v-1 an eigenvalue of A-1?
c. Is v + v-1 an eigenvalue for A + A-1?
d. What about (v + v-1)³ ?
6 Given A = PDP-1, find E and Q such that A + A² + A³ = Q-1EQ. If appropriate for you, use the matrices A and P from Q3.
7 If C is the 3x3 matrix with rows (1 2 3) (0 2 7/2) (-3 6 0) then show that the latent roots (another term for eigenvalues) are integers a,b,c such that a+2b+c=0 and find the matrices D and P so that C = PDP-1.
8 Does it matter whether you do post-multiplication, (PQ)R or pre-multiplication, P(QR)?
Generally one is supposed to do pre-multiplication (“do this”, verb first). Assume D is a diagonal matrix, that A is a more general matrix and P is the matrix constructed from the eigenvectors of A. Consider the eight combinations (PA)P-1, P-1(AP), P(AP-1), (P-1A)P, (PD)P-1, (P-1D)P, P(DP-1), P-1(DP). They fall into pairs and these pairs equate to A, to D or to Q or R (different, less obviously related matrices). Identify which pair equates to which result and try to explain why this is so. If necessary, use the matrix A from Q3. You may wish to prove your answer to the first line.
Answers? You jest....