Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I ran into the following problem and its solution:

The integration by parts formula $$ \int_{a}^{b}u\frac{dv}{dx}\,dx=uv\bigg|_{a}^{b}-\int_{a}^{b}v\frac{du}{dx}\,dx $$ is known to be valid for functions $u(x)$ and $v(x)$, which are continuous and have continuous first derivatives. However, we will assume that $u$, $v$, $du/dx$, and $dv/dx$ are continuous only for $a\leqslant x\leqslant c$ and $c\leqslant x \leqslant b$; we assume that all quantities may have a jump discontinuity at $x=c$.

(a) Derive an expression for $\int_{a}^{b}u\,dv/dx\,dx$ in terms of $\int_{a}^{b}v\,du/dx\,dx$. $$ \int_{a}^{b}u\frac{dv}{dx}\,dx=uv\bigg|_{a}^{b}+uv\bigg|_{c^+}^{c^-}-\int_{a}^{b}v\frac{du}{dx}\,dx. $$

Could anyone clarify to me how this was obtained?

Edit: Following the advice of Muphrid, I obtained the following:

$$\begin{align} \int_{a}^{c^-}u\frac{dv}{dx}\,dx+\int_{c^+}^{b}u\frac{dv}{dx}\,dx&=uv\bigg|_{a}^{c^-}-\int_{a}^{c^-}v\frac{du}{dx}\,dx+uv\bigg|_{c^+}^{b}-\int_{c^+}^{b}v\frac{du}{dx}\,dx,\\ \int_{a}^{b}u\frac{dv}{dx}\,dx&=\color{red}{uv\bigg|_{a}^{c^-}+uv\bigg|_{c^+}^{b}}-\int_{a}^{b}v\frac{du}{dx}\,dx. \end{align}$$

What is the rule for combining the terms in red?

share|cite|improve this question
up vote 5 down vote accepted

In response to your work from Muphrid:

$$\begin{align} \color{red}{uv\bigg|_{a}^{c^-}+uv\bigg|_{c^+}^{b}} &= uv(c^-) - uv(a) + uv(b) - uv(c^+) \\ &= uv(c^-) - uv(c^+) + uv(b) - uv(a) \\ &= uv\bigg|_{c^+}^{c^-} +uv\bigg|^b_a \end{align}$$

share|cite|improve this answer
I am sorry; I changed the limits of integration after you saw my edit, but your derivation follows exactly the same way. Thank you! – EsX_Raptor Nov 2 '12 at 22:12
I see that now, and went ahead and fixed my answer. Best of luck. – mixedmath Nov 2 '12 at 22:14

I suggest using the usual integration by parts formula on the separate intervals $[a,c)$ and $(c, b]$ and then putting the two results together.

share|cite|improve this answer

No matter how you derive it, this formula is wrong. Consider the example $$ \int_{0}^2 x \, H(x-1) \; dx, $$ where $H(x)$ is the unit step function. Modifying the limits to account for $H(x-1)$ gives $$ \int_{0}^2 x \, H(x-1) \; dx= \int_{1}^2 x \; dx = \frac{4}{2} - \frac{1}{2} = \frac{3}{2} $$

Applying standard integration by parts with $v = \frac{x^2}{2}$ and $u = H(x-1)$ gives the correct answer $$ \int_{0}^2 x \, H(x-1) \; dx= \frac{x^2}{2} H(x-1) \bigg|_{0}^2 - \int_{0}^2 \frac{x^2}{2} \delta(x-1) \; dx = \frac{4}{2} - 0 - \frac{1}{2} = \frac{3}{2}. $$

Since $u$ has a jump at $c=1$, the modified result proposed this question has the additional non-zero term $$ \frac{x^2}{2} H(x-1) \bigg|_{1^-}^{1^+} = \frac{1}{2}. $$ Thus, the modified formula is not correct.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.