linear algebra
posted by mathstudent .
if:
A and B are matrices
and A^2 is similar to B^2
Is A guaranteed to be similar to B?

Matrix similarity means that the matrices are identical if one of the matrices is converted to another basis. If matrices C and D are similar:
C = P^1 * D * P
where P converts from standard basis to alternate basis B, and P^1 is the inverse.
Counterexample: take A to be the identity matrix and define B by changing a 1 to 1 somewhere on the diagonal. Then A^2 = B^2, but A is not equal to B.
No transformation of the form
P A P^(1) can make A equal to B. If two matrices are similar and diagonizable, then both matrices can be obtained from the same diagonal matrix using different transformaton matrices.
So, in diagonalized form they must be the same. But B and A are already in diagonalized form and are not the same.
great counterexample. thanks count.
Respond to this Question
Similar Questions

Math (matrices)
No one answered my matrix question. Let me rephrase: Let A, B, and 0 be 2x2 matrices. Assuming that A is invertible and 0 is all zeroes, what is the inverse of the matrix [A0] [BA] (that is a 4x4 matrix represented as 4 2x2 matrices) … 
Math: matrices
If A and B are both square n x n matrices, If AB = I, prove BA = I Presumably you have to do this without using the usual properties of the inverse of matrices. But we do need to use that if there exists a matrix B such that A B = … 
linear algebra
Prove that the trace is a similarity invariant. In other words, if two matrices are similar, then they must have the same trace. Got the answer from Wikipedia: tr(AB) = tr(BA) tr(ABC) = tr(CAB) tr(P^1 * A * P) = tr(P^1 * P * A) = … 
Algebra II
Can a matrix have a two digit number? I have to add the matrices[3 4] +[2 7] [7 6] +[3 8] Sorry this is the only way I know haw to put matrices on the board. 
Algebra II (Matrices)
My book doesn't solve it like a linear equation, they solve it by using inverse matrices. Solve the matrix equation: 4 5m=32 1 2 n=5 A= 4 5 1 2 X= m n B= 32 5 Step 1. Find the inverse of the coefficient … 
matrices
Two matrices can be multiplied only if their sizes are compatible. Suppose that U is an m × n matrix, and that V is a p × q matrix. In order for U•V to make sense, what must be true of the dimensions of these matrices? 
math
Two matrices can be multiplied only if their sizes are compatible. Suppose that U is an m × n matrix, and that V is a p × q matrix. In order for U•V to make sense, what must be true of the dimensions of these matrices? 
Math
I have a few questions about TMatrix. In excel, I am suppose to work with powered matrices to construct a weighted T matrix, using a scalar of .7. Does this mean I multiply each of the powered matrices by .7? 
Linear Algebra
Diagonalize the matrix A = [1,1;2,4] using a similarity transformation. Give the transformation matrices such that (C^−1)AC = D, where D is a diagonal matrix. (D, C) = ? 
Linear Algebra
Diagonalize the matrix A = [1,1;2,4] using a similarity transformation. Give the transformation matrices such that (C^−1)AC = D, where D is a diagonal matrix. (D, C) = ?