I am tring to solve the following task:

Linear transformation $f: \mathbb R^2 \rightarrow \mathbb R^2$ is given by $f(\begin{bmatrix} x_1\\ x_2 \end{bmatrix}) = \begin{bmatrix} 2x_1-x_2\\ x_1+x_2 \end{bmatrix}$. Answer true or false to the following questions:

a) in some basis of transformation $ \mathbb R^2$ transformation matrix of $f$ is $\begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix}$

b) $f$ is a bijection

c) transformation matrix of $f$ in basis $([1,0]^T, [1,1]^T)$ is $\begin{bmatrix} 1 & 1 \\ -1 & 2 \end{bmatrix}$

Could you kindly give me any HINTS how to start it (not solution)?

  • $\begingroup$ a) How are the matrices of a linear transformation in different bases related. b) What criterion have you for a linear transformation to be a bijection? c) Calculate $f((1,0))$ and $f((1,1))$ and see whether the results match. $\endgroup$ – Daniel Fischer Jun 28 '13 at 14:02

Hints (assuming some familiarity with various results of linear algebra):

a) If $A$ and $B$ are transformation matrices of $f$ with respect to different bases, then there is some matrix $X$ such that $XAX^{-1} = B$. What does this say if $A$ is the identity matrix?

b) A linear transformation is bijective if and only if for any (and thus all) basis, its transformation matrix is invertible.

c) This is a calculation.

  • $\begingroup$ +1 Ok, thanks a lot for your help! I have obtained False, True, False. But b) point I have proved not by criterion with basis, but checking conditions on suriection and injection. Could you explain me how to prove it using your criterion (with basis and invertible matrix)? $\endgroup$ – JosephConrad Jun 28 '13 at 15:28
  • $\begingroup$ Do you believe that the criterion is true? If so, in the standard basis, the matrix for $f$ is $\pmatrix{ 2 & -1 \\ 1 & 1 }$. Now, there are tons of criteria for a matrix being invertible, typically the easiest one to check being that the matrix has non-zero determinant, which is definitely the case for the matrix here. $\endgroup$ – fuglede Jun 28 '13 at 21:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.