Advanced Linear Algebra
Math 424/524

Assignment No. 2

Due October 16, 2002

  1. Prove that if two matrices of the same size over a field are both in reduced row echelon form and are also row equivalent, i.e., each may be obtained from the other by a finite sequence of elementary row operations, then they must be equal.

  2. Let R denote the field of real numbers.

    1. What is the linear subspace of R^{n} spanned by the set of columns x \in R^{n} having the property that every coordinate of x is non-zero?

    2. What is the linear subspace of the vector space M_{n}(R) of n \times n real matrices that is spanned by the set of invertible n \times n matrices?

  3. Let F be a field, and let V denote the vector space F[t] of polynomials in the variable t with coefficients in F. Let a be a given element of F, and let U_{a} be the linear subspace of V consisting of all polynomials divisible by the polynomial t - a. Does the isomophism class of the quotient space V/U_{a} depend on the choice of a ?

    Hint: Consider the linear map s_{a}: F[t] --> F defined by s_{a}(f) = f(a).

  4. If F is any field, let V be the vector space M_{n}(F) of n \times n matrices over F. For given A, B \in V define an F-linear endomorphism phi_{A,B} of V by

     phi_{A,B}(M)  =  A M B  .  
    1. For what pairs A, B is the endomorphism phi_{A,B} equal to 0 ?

    2. Is every endomorphism of V equal to phi_{A,B} for some pair A, B ?

  5. Let X be a vector space over a field F, and let psi be a linear map from X to X for which

     psi \circ  psi  =  psi  .  
    Let V be the subspace of X that is the image of psi, let j be the inclusion of V in X, and let q: X --> V be the “projection” of X on V that yields the canonical factorization psi = j \circ q of psi through its image.

    Define a subspace U of X and a linear map p: X --> U so that, with i: U --> X the inclusion of U in X, one has the relations among i, j, p, and q characterizing an isomorphism of X with the Cartesian product U \times V, i.e.,

    p i 
     = 
    1
    p j
     = 
    0
    q i
     = 
    0
    q j
     = 
    1
    i p + j q
     = 
    1


AUTHOR  |  COMMENT