Term-by-term operations on series of functions

Uniform convergence hypotheses justify integrating or differentiating a function series term-by-term
Term-by-term operations on series of functions

Let n=1fn\sum_{n=1}^\infty f_n be a series of functions on an [a,b][a,b].

Proposition (term-by-term integration): Suppose each fnf_n is on [a,b][a,b] and fn\sum f_n on [a,b][a,b] to a function ff. Then ff is Riemann integrable and abf(x)dx=n=1abfn(x)dx. \int_a^b f(x)\,dx=\sum_{n=1}^\infty \int_a^b f_n(x)\,dx.

Proposition (term-by-term differentiation): Suppose each fnf_n is on [a,b][a,b], and:

  • the series of fn\sum f_n' converges uniformly on [a,b][a,b] to a function gg, and
  • the original series fn(x0)\sum f_n(x_0) at some point x0[a,b]x_0\in[a,b]. Then fn\sum f_n converges uniformly on [a,b][a,b] to a differentiable function ff, and f(x)=n=1fn(x)for all x[a,b]. f'(x)=\sum_{n=1}^\infty f_n'(x)\quad\text{for all }x\in[a,b].

These statements formalize the usual calculus manipulations with function series; the uniform convergence hypotheses are the key analytic input.

Proof sketch: Integration: apply the “ ” theorem to partial sums sN=n=1Nfns_N=\sum_{n=1}^N f_n. Differentiation: apply the “ ” theorem to the sequence of partial sums sNs_N, noting sN=n=1Nfns_N'= \sum_{n=1}^N f_n' and using the convergence at x0x_0 to pin down constants of integration.