Page 276 - DMTH401_REAL ANALYSIS
P. 276

Real Analysis




                    Notes          22.1 Riemann-Stieltjes Sums

                                   A Riemann-Stieltjes sum for a function f(x) defined on an interval [a, b] is formed with the help
                                   of

                                   1.  A partition  of [a, b], namely an ordered, finite set of points x, with a = x  < x  < < x  = b
                                                                                        i        0  1      n
                                       (where n is a positive integer that can be any positive integer, and one that we will often
                                       write as n = n ),
                                                  
                                   2.  A selection vector  = ( , ...,  ) that has n  components that must satisfy x      x , for
                                                           1    n                                 i – 1   i  i
                                       i = 1, 2, ..., n.
                                       and
                                   3.  An integrator (x), which is a function defined on [a, b] that plays the role of the x in dx ...
                                       A Riemann-Stieltjes sum for f over [a, b] with respect to the partition , using the selection
                                       vector , and integrator , may be denoted (in greatest detail!) as follows, and it is given
                                       by the value of the sum following it:

                                                                n 
                                   4.         RS (f, , [a, b], , ):= å  f( )( (x ) -  (x i 1  )) .
                                                                   
                                                                      
                                                                        i
                                                                    i
                                                                               -
                                                                =
                                                                i 1
                                   22.2 More Notation: The Mesh (Size) of a Partition
                                   In this definition, as in the Riemann-sums definition, we can write x := x —x   or  := (x ) –
                                                                                          i   i  i – 1  i    i
                                   (x  ). These are convenient because they are short and suggest the dx or d in an integral. But
                                     i – 1
                                   they can cause confusion because they leave out the dependence they have on x  . The x  is used
                                                                                                 i – 1   i
                                   in the Riemann-Stieltjes context.
                                   A partition  can be thought of as “dividing” the interval [a, b] into subintervals. We may write
                                   |[a, b] and read this as “ divides [a, b],” or “partitions [a, b].” We will denote the intervals of
                                    by I := [x  , x ]. When we wish to work with 2 partitions at the same time we will have to
                                       i   i – 1  i
                                   distinguish between them somehow, for example we can use y  to denote the other’s points and
                                                                                     j
                                   J to denote its intervals, etc.
                                   j
                                   We measure the fineness of a partition using the length of the longest interval in the partition.
                                   This number is written
                                                       mesh():= max (x – x  ) =  max x .
                                                                 
                                                                               
                                                                1 i n   i   i – 1  1 i n   i
                                   This definition of mesh size is used and not  max ((x ) –(x  )) even in the Riemann-Stieltjes
                                                                         
                                                                       1 i n   i   i – 1
                                   context.
                                   22.3 The Riemann-Stieltjes – sum Definition of the
                                        Riemann-Stieltjes Integral
                                   Definition: A real-valued function f(x) defined on the bounded and closed interval [a, b]  is
                                   Riemann-Stieltjes integrable on [a, b] with respect to  if there exists a number RSI such that for
                                   all  > 0 there exists  > 0 such that for every partition  of [a, b],
                                                   mesh() <   |RS(f, ) – RSI| < .

                                   We write
                                                         b      b
                                                                      
                                                         a ò  f d = ò a  f(x) d (x) := RSI




          270                               LOVELY PROFESSIONAL UNIVERSITY
   271   272   273   274   275   276   277   278   279   280   281