2

I noticed in a proof that he needed to state that one integral approached another:

$$ \lim_{a \to \infty} \int_0^\infty f(x-a) \, dx = \int_{-\infty}^\infty f(x) \, dx $$

Does this always hold for integrable functions (Riemann or Lebesgue) functions $f \in L^1(\mathbb{R})$ ? This seems rather obvious.

$$ \int_{-\infty}^\infty f(x) \, dx - \int_0^\infty f(x-a) \, dx =\int_{-\infty}^\infty f(x) \, dx - \int_{-a}^\infty f(x) \, dx =\int_{-\infty}^{-a} f(x) \, dx \stackrel{?}{\to} 0 $$

Here I would have used the linearity of the integral. Doesn't the integral on the right-hand-side tend to zero?

cactus314
  • 24,438
  • in light of the answer: https://math.stackexchange.com/questions/1779831/why-isnt-dominated-convergence-theorem-taught-in-intro-analysis – cactus314 Sep 10 '18 at 23:15

1 Answers1

1

Yes, this is correct. The convergence of $\int_{-\infty}^{-a} f(x) \, dx$ to $0$ follows from the dominated convergence theorem, for instance (we are integrating the functions $f\cdot 1_{(-\infty,-a)}$ which are dominated by the integrable function $|f|$).

Eric Wofsey
  • 330,363