Introduction
When a textbook states that the functions f and g are integrable, it opens the door to a rich collection of analytical tools that make it possible to manipulate, combine, and extract meaningful information from these functions. Integrability—whether in the Riemann, Lebesgue, or improper sense—guarantees that the area under each curve can be measured in a precise way. This article explores the consequences of having two integrable functions, focusing on the algebra of integrable functions, convergence theorems, common pitfalls, and practical examples that illustrate why integrability matters in pure and applied mathematics.
1. What Does “Integrable” Mean?
1.1 Riemann integrability
A bounded function (f:[a,b]\to\mathbb{R}) is Riemann integrable if the upper and lower Darboux sums converge to the same limit as the partition mesh tends to zero. In practice, this is equivalent to the statement that the set of discontinuities of (f) has Lebesgue measure zero (the Lebesgue criterion).
1.2 Lebesgue integrability
A measurable function (f:\Omega\to\mathbb{R}) (or (\mathbb{C})) is Lebesgue integrable on a measurable set (\Omega) if
[ \int_{\Omega} |f|,d\mu < \infty . ]
Here the absolute value ensures that the integral of the positive and negative parts are both finite, which is a stricter requirement than merely having a finite improper integral.
1.3 Improper integrability
When the domain is unbounded or the function has singularities, we speak of improper integrability:
[ \int_{a}^{\infty} f(x),dx = \lim_{R\to\infty}\int_{a}^{R} f(x),dx, ]
provided the limit exists and is finite.
Understanding which notion of integrability applies is essential because the algebraic properties we will discuss hold under specific frameworks.
2. Basic Algebraic Properties
If f and g are integrable on the same interval (or measurable set), the following operations preserve integrability:
| Operation | Resulting Function | Condition |
|---|---|---|
| Sum | (f+g) | Always integrable (Riemann or Lebesgue) |
| Difference | (f-g) | Always integrable |
| Scalar multiple | (c,f) (with (c\in\mathbb{R})) | Always integrable |
| Product | (f\cdot g) | Lebesgue integrable if (f,g\in L^{2}) (or more generally (f\in L^{p}, g\in L^{q}) with (1/p+1/q=1)). For Riemann integrability, the product of two bounded Riemann‑integrable functions is again Riemann‑integrable. Still, |
| Quotient | (\frac{f}{g}) | Integrable on the set where (g\neq0) and (\frac{1}{g}) is bounded; otherwise additional hypotheses are needed. |
| Maximum / Minimum | (\max{f,g},; \min{f,g}) | Integrable because they can be expressed using sums and absolute values: (\max{f,g}=\frac{f+g+ |
These properties are not merely formal; they make it possible to construct new integrable functions from known ones, which is a cornerstone of functional analysis, probability theory, and numerical approximation.
3. Integral Linearities and Inequalities
3.1 Linearity
If f and g are integrable on ([a,b]), then for any constants (\alpha,\beta),
[ \int_{a}^{b} (\alpha f + \beta g) ,dx = \alpha\int_{a}^{b} f,dx + \beta\int_{a}^{b} g,dx . ]
This property follows directly from the definition of the integral and is the basis for Fourier series, Laplace transforms, and many other linear operators Turns out it matters..
3.2 Triangle inequality
[ \Bigl|\int_{a}^{b} f(x),dx\Bigr| \le \int_{a}^{b} |f(x)|,dx . ]
This means if f and g are integrable, then
[ \Bigl|\int_{a}^{b} (f+g),dx\Bigr| \le \int_{a}^{b} |f|,dx + \int_{a}^{b} |g|,dx . ]
These bounds are useful in error analysis for numerical integration and in proving convergence of sequences of integrals Worth knowing..
4. Convergence Theorems Involving Two Integrable Functions
When dealing with sequences ({f_n}) and ({g_n}) that converge to f and g, respectively, integrability of the limits is often guaranteed by powerful convergence theorems Simple, but easy to overlook..
4.1 Dominated Convergence Theorem (DCT)
If (|f_n|\le h) and (|g_n|\le h) for an integrable dominating function (h), and (f_n\to f), (g_n\to g) pointwise a.e., then
[ \lim_{n\to\infty}\int (f_n+g_n) = \int (f+g). ]
The DCT shows that the sum of the limits is the limit of the sums, preserving integrability.
4.2 Monotone Convergence Theorem (MCT)
If (0\le f_n\uparrow f) and (0\le g_n\uparrow g), then
[ \int (f+g) = \lim_{n\to\infty} \int (f_n+g_n) . ]
Both theorems underline the stability of integrability under limits, a fact that is essential when constructing functions piecewise.
5. Practical Examples
Example 1: Sum of Two Riemann‑Integrable Functions
Let
[ f(x)=\sin x,\qquad g(x)=\begin{cases} 1 & \text{if } x\in\mathbb{Q}\cap[0,1],\[2pt] 0 & \text{otherwise}. \end{cases} ]
Both are bounded on ([0,1]). (f) is continuous, hence Riemann integrable. (g) is the characteristic function of the rationals, whose set of discontinuities is the whole interval, but because the rationals have measure zero, (g) is Riemann integrable with integral (0) Worth knowing..
[ \int_{0}^{1} h(x),dx = \int_{0}^{1}\sin x,dx + \int_{0}^{1} g(x),dx = 1-\cos 1 + 0 . ]
Example 2: Product in the Lebesgue Sense
Take (f(x)=\frac{1}{\sqrt{x}}) on ((0,1]) and (g(x)=\sqrt{x}) on the same interval. Both belong to (L^{2}(0,1)) because
[ \int_{0}^{1} |f|^{2}dx = \int_{0}^{1} \frac{1}{x},dx = \infty, ]
so actually (f\notin L^{2}). Even so, (f\in L^{1}) and (g\in L^{\infty}). Their product (f\cdot g = 1) is trivially integrable with
[ \int_{0}^{1} 1,dx = 1 . ]
This illustrates that integrability of a product does not require both factors to be square‑integrable; a bounded factor suffices.
Example 3: Improper Integral of a Quotient
Let (f(x)=\frac{\sin x}{x}) and (g(x)=\frac{1}{x^{2}+1}) on ([1,\infty)). Both are improper integrable because
[ \int_{1}^{\infty} \frac{|\sin x|}{x},dx < \infty,\qquad \int_{1}^{\infty} \frac{1}{x^{2}+1},dx = \frac{\pi}{2}-\arctan 1 . ]
The quotient (\frac{f}{g}= \sin x ,\frac{x^{2}+1}{x}) behaves like (\sin x,x) for large (x), which is not integrable. In practice, hence, while each function is integrable, the quotient may fail to be. This example warns against assuming that all algebraic combinations preserve integrability without checking growth conditions That's the part that actually makes a difference..
No fluff here — just what actually works.
6. Frequently Asked Questions
Q1. If f and g are Riemann integrable, is their absolute value (|f-g|) also integrable?
A: Yes. The absolute value of a Riemann‑integrable function is integrable because (|f-g|) can be expressed as (\sqrt{(f-g)^{2}}), and the square of an integrable function is integrable; taking the square root preserves integrability on bounded intervals.
Q2. Does integrability imply continuity?
A: No. Integrability only requires that the set of discontinuities have measure zero (Riemann) or that the function be measurable with finite integral (Lebesgue). Functions such as the Dirichlet function on rationals are not integrable, but the characteristic function of the Cantor set is Lebesgue integrable despite being discontinuous everywhere on that set.
Q3. Can we integrate a series term‑by‑term if each term is integrable?
A: Under uniform convergence or when a dominating integrable function exists (DCT), term‑by‑term integration is justified. Otherwise, the series of integrals may converge while the integral of the sum diverges.
Q4. Is the set of integrable functions a vector space?
A: Yes. Both the Riemann‑integrable and Lebesgue‑integrable functions on a fixed domain form vector spaces over (\mathbb{R}) (or (\mathbb{C})) because the sum and scalar multiples of integrable functions remain integrable.
Q5. How does integrability relate to probability?
A: In probability theory, random variables are measurable functions. A random variable (X) is integrable (i.e., has a finite expectation) precisely when (\mathbb{E}[|X|] < \infty). The linearity of expectation is a direct consequence of the linearity of the Lebesgue integral.
7. Common Pitfalls and How to Avoid Them
-
Assuming product integrability without checking growth.
Solution: Verify that at least one factor is essentially bounded or belongs to a suitable (L^{p}) space so that Hölder’s inequality applies. -
Confusing pointwise convergence with integrable convergence.
Solution: Use DCT or MCT to justify interchanging limit and integral; otherwise, construct a counterexample (e.g., (f_n(x)=n\mathbf{1}_{[0,1/n]})). -
Neglecting the domain of definition for quotients.
Solution: Identify the set where the denominator vanishes, and check that its measure is zero or that the reciprocal is bounded on the complement Still holds up.. -
Treating improper integrals as automatically convergent.
Solution: Apply comparison tests (e.g., limit comparison with (1/x^{p})) to confirm convergence Less friction, more output..
8. Conclusion
The statement “f and g are integrable” is a powerful starting point for a wide array of mathematical operations. Linearities and inequalities give us control over the size of integrals, while convergence theorems protect us when passing to limits. Integrability guarantees that we can add, subtract, and often multiply these functions while preserving the ability to compute meaningful “areas” or expectations. Even so, integrability is not an all‑encompassing safety net—products, quotients, and series require careful verification of additional conditions No workaround needed..
By mastering the algebraic and analytical consequences of integrability, students and practitioners can confidently manipulate functions in calculus, analysis, probability, and physics, knowing exactly when the tools of integration remain valid. Whether you are estimating an integral numerically, proving a theorem in functional analysis, or modeling a stochastic process, the integrability of f and g is the foundation upon which rigorous, reliable results are built.
