Dear Sir/Madam: I try the following maple commands: > with(linalg): > A:=matrix([[r1*X, (1-r1)*X],[1-r2,r2]]): > eigsys:=eigenvectors(A); > ES[1]:=eigsys[1][3];ES[2]:=eigsys[2][3]; > P:=augment(op(ES[1]),op(ES[2])); > K:=map(simplify,evalm(inverse(P)&*A&*P)); > W:=map(simplify,evalm(P&*K&*inverse(P))); > U:=map(expand, W); > map(simplify,U); Now, K and P are correct because P&*K&*inverse(P) = A. But, when I use the following maple commands, I cannot get A^i with entries that are polynomials of r1 and r2 and X. > V:=map(simplify, evalm(P&*(K^i)&*inverse(P))); (Note: formula A^i = P*K^i* inverse of P is correct.)

Dear Sir/Madam:

I try to find a close form for A^k,
where k is an arbitrary interger and A is 2 by 2 matrix as the following:
A = ([[r1*X, (1-r1)*X],[1-r2,r2]]);
r1 and r2 are symbolic constants, X is a variable.

First, I run the following commands in maple:

A:=matrix([[r1*X, (1-r1)*X],[1-r2,r2]]);
Z3:=A^3;
simplify(evalm(Z3),symbolic);

I hope to simplyfy the entries of Z3 in the order of X, i.e.,

z11 = a0*X^3 + a1*X^2 + a2*X +a3;
z12 = ....... (the same as the above)
z21 = .......
z22 = .......

But I do not know how to do it.

I try to find a law for representations of the entries of A^k by looking at the cases k=3,4,....