Skew-Symmetric Matrix proof

RavenCpu

Hello all! I finally decided to join this forum after quite a long time of lurking haha. Anywho, to the point. A friend of mine is taking linear algebra and is having a lot of issues with the below posted problem. I'm trying to help her with it, but sadly I'm having issues with it as well haha. I would appreciate any help you all could offer me in doing this proof. I've gotten it all solved up until I have to prove that [itex]M_{n\times n}(F) = W_1 \oplus W_2[/itex]. So basically, I'm on the last step.

Prove:
A matrix M is called a skew-symmetric if [itex]M^t = -M[/itex]. Clearly, a skew-symmetric matrix is square. Let F be a field. Prove that the set [itex]W_1[/itex] of all skew-symmetric n x n matrices with entries from F is a Subspace of [itex]M_{n \times n}(F)[/itex]. Now assume that F is not of characteristic 2, and let [itex]W_2[/itex] be the subspace of [itex]M_{n \times n}(F)[/itex] conisting of all symmetric n x n matrices. Prove that [itex]M_{n\times n}(F) = W_1 \oplus W_2[/itex]

AKG

Rewriting it so it's legible:

Hello all! I finally decided to join this forum after quite a long time of lurking haha. Anywho, to the point. A friend of mine is taking linear algebra and is having a lot of issues with the below posted problem. I'm trying to help her with it, but sadly I'm having issues with it as well haha. I would appreciate any help you all could offer me in doing this proof. I've gotten it all solved up until I have to prove that [itex]M_{n\times n}(F) = W_1 \oplus W_2[/itex]. So basically, I'm on the last step.

Prove:
A matrix M is called a skew-symmetric if [itex]M^t = -M[/itex]. Clearly, a skew-symmetric matrix is square. Let F be a field. Prove that the set [itex]W_1[/itex] of all skew-symmetric n x n matrices with entries from F is a Subspace of [itex]M_{n \times n}(F)[/itex]. Now assume that F is not of characteristic 2, and let [itex]W_2[/itex] be the subspace of [itex]M_{n \times n}(F)[/itex] conisting of all symmetric n x n matrices. Prove that [itex]M_{n\times n}(F) = W_1 \oplus W_2[/itex]

AKG

1. Prove that [itex]W_1,\, W_2[/itex] are subspaces
2. Prove [itex]W_1 \cap W_2 = \{ 0\}[/itex]
3'. Find bases [itex]\beta _1,\, \beta _2[/itex] and show that [itex]\beta _1 \cup \beta _2[/itex] is a basis for [itex]M_{n\times n}(F)[/itex]. This is easy since it's just a matter of comparing [itex]|\beta _1| + |\beta _2|[/itex] to dim(M_{n x n}(F)).
3''. Instead of 3', you could do the following: given any square matrix A, use your bases to compute the projection of A onto each of those subspaces, and show that the sum of the two projections is A.
3'''. You should do 3' or 3'', but another way is to prove that for any given matrix A, the equation A = (1/2)(A + A^t) + (1/2)(A - A^t) holds, and that a) A + A^t is symmetric, b) A - A^t is skew symmetric, and c) (1/2) makes sense because F is not of characteristic 2. This is the most efficient way to prove it, but it requires you to know how to decompose A in the first place. Doing the first two ways will actually teach you how, in the future, you can go about proving that some vector space is the direct sum of some subspaces. This third approach is just a handy tid-bit of knowledge, not a very instructive approach though.

RavenCpu

Thanks for rewriting it :-) I'm not too familiar with symbols and the sort in forums. It looks so clean now!

Thank you for the help! I'll give it an attempt and see what comes of it.