Page 88 - DMTH402_COMPLEX_ANALYSIS_AND_DIFFERENTIAL_GEOMETRY
P. 88

Unit 8: Series




                                                                                                Notes
                                 
                              
                                    
                            <         , 
                              3  3  3
          Now suppose, we have a sequence (f ) of continuous functions which converges uniformly on a
                                        n
                                                      
                                                  n 
                                                                     n 
          contour C to the function f. Then the sequence     f (z)dz   converges to  f (z)dz.  This is easy to
                                                       
                                                C                C
                                                  
          see. Let  > 0. Now let N be so that |f (z) – f(z)| <  A   for n > N, where A is the length of C. Then,
                                       n
               n 
                                    
              f (z)dz   f(z)dz  =   (f (z) f(z))dz
                                 n
             C       C         C
                            <    A  
                              A
          whenever n > N.
          Now suppose (f ) is a sequence of functions each analytic on some region D, and suppose the
                       n
          sequence converges uniformly on D to the function f. Then f is analytic. This result is in marked
          contrast to what happens with real functions—examples of uniformly convergent sequences of
          differentiable functions with a non-differentiable limit abound in the real case. To see that this
          uniform limit is analytic, let z   D, and let S = {z : |z – z | <  r}  D. Now consider any simple
                                  0
                                                        0
                                                           0
          closed curve C   S. Each  f   is analytic,  and so  f (z)dz    for every n.  From the  uniform
                                                    n 
                                 n
                                                  C
                                                                             
                                        
                                                                         n 
          convergence of  (f ), we  know that  f(z)dz   is the  limit of  the sequence     f (z)dz ,    and  so
                                                                              
                         n
                                        C                               C    
             f(z)dz  0.  Morera’s theorem now tells us that f is analytic on S, and hence at z . Truly a miracle.
                                                                         0
          C
          8.2 Series
          A series is simply a sequence (s ) in which s  = a  + a  + ... + a . In other words, there is sequence
                                                    2
                                                1
                                                           n
                                             n
                                   n
          (a ) so that s  = s n – 1  + a . The s  are usually called the partial sums. Recall from Mrs. Turner’s class
                    n
                            n
            n
                                 n
                          n  
          that if the series   a j   has a limit, then it must be true that  lim(a ) 0.
                                                          n  n
                          j 1  
                         
                           n  
          Consider a series   fj(z)  of functions. Chances are this series will converge for some values
                              
                           j 1  
                          
          of z and not converge for others. A useful result is the celebrated Weierstrass M-test: Suppose
          (M) is a sequence of real numbers such that M  0 for all j > J, where J is some number., and
                                                j
            j
                                   n  
          suppose also that the series   M j   converges. If for all z  D, we have |fj(z)|  M for all j > J,
                                                                           j
                                   j 1  
                                  
                         n  
          then the series   f (z)  converges uniformly on D.
                         j  
                         j 1  
                        
                                           LOVELY PROFESSIONAL UNIVERSITY                                   81
   83   84   85   86   87   88   89   90   91   92   93