Page 203 - DMTH502_LINEAR_ALGEBRA
P. 203

Unit 17: Invariant Direct Sums




          If T is diagonalizable, T = c E  + ... + c E , then                                   Notes
                                1  1    k  k
                                      g(T) = g(c )E  + ... + g(c )E
                                              1  1      k  k
          for every polynomial g. Thus g(T) = 0 if and only if g(c ) = 0 for each i. In particular, the minimal
                                                     i
          polynomial for T is
                                         p = (x – c ) ... (x – c )
                                                1      k
          Now suppose T is a linear operator with minimal polynomial p = (x – c ) ... (x – c ), where c ,..., c
                                                                  1       k       1   k
          are distinct elements of the scalar field. We form the Lagrange polynomials
                                                (x c  )
                                                  
                                           j 
                                          p  =      i
                                              i j  (c   c  i )
                                              
                                                  j
          So that p (c ) =   and for any polynomial g of degree less than or equal to (k – 1) we have
                 j  i  ij
                                       g = g(c )p  + ... + g(c )p
                                             1  1      k  k
          Taking g to be the scalar polynomial 1 and then the polynomial x, we have
                                       1   p     p  
                                           1      k                                ...(2)
                                       x   c p    c p k 
                                                   k 
                                           1 1
          You will note that the application to x may not be valid because k may be 1. But if k = 1, T is a
          scalar multiple of the identity and hence diagonalizable). Now let E  = p (T). From (2) we have
                                                                 j  j
                                       I   E    E k  
                                           1
                                                                                  ...(3)
                                       T   c E   c E
                                           1  1    k  k 
          Observe that if i  j, then p p  is divisible by the minimal polynomial p, because p p  contains
                                i   j                                        i  j
          every (x – c ) as a factor. Thus
                   r
                                  E E  = 0,  i  j                                 ...(4)
                                   i  j
          We must note one further thing, namely, that E   0 for each i. This is because p is the minimal
                                                i
          polynomial for T and so we cannot have p (T) = 0 since p  has degree less than the degree of p. This
                                           i          i
          last comment, together with (3), (4), and the fact that the  c  are distinct  enables us to apply
                                                           i
          Theorem 2 to conclude that T is diagonalizable.
          Self Assessment
          1.   Let T be the diagonalizable linear operator on R  which is represented by the matrix
                                                      3
                     5   6  6
                            
               A   1  4  2
                            
                     3   6  4
                            
               use the Lagrange polynomials to write the representing matrix A in the form A = E  + 2E ,
                                                                                 1   2
               E  + E  = I, E E  = 0. Where I is a unit matrix and 0 is zero matrix.
                1   2    1  2
                                         4
          2.   Let T be the linear operator on R  which is represented by the 4 × 4 matrix
                    0 1 0 1
                    1 0 1 0 
               A          
                    0 1 0 1
                           
                    1 0 1 0 
               Find the matrices E , E , E  such that
                              1  2  3
               A = C E  + C E  + C E , E  + E  + E  = I and E E  = 0 for i  j
                   1  1  2  2  3  3  1  2  3      i  j


                                           LOVELY PROFESSIONAL UNIVERSITY                                   197
   198   199   200   201   202   203   204   205   206   207   208