My Math Forum Determine the sign of the following integral without integrating it

 Real Analysis Real Analysis Math Forum

 December 13th, 2017, 10:31 AM #1 Member   Joined: Jan 2016 From: Blackpool Posts: 98 Thanks: 2 Determine the sign of the following integral without integrating it Determine the sign of the following integral without integrating it: $\int_{0}^{\pi}x{\cos(x)}dx$ Last edited by skipjack; December 13th, 2017 at 12:13 PM.
 December 13th, 2017, 10:48 AM #2 Senior Member   Joined: Sep 2016 From: USA Posts: 475 Thanks: 262 Math Focus: Dynamical systems, analytic function theory, numerics I'm not even sure how to give a hint here. If you know what an integral is, then this question should be completely trivial. Maybe try computing this integral (despite the instructions) so you know the answer, and then try to figure out why this answer should have been easy to see without computation. Thanks from Jaket1
 December 13th, 2017, 11:34 AM #3 Member   Joined: Jan 2016 From: Blackpool Posts: 98 Thanks: 2 My theory is that it will be negative because at x=pi we have pi*cos(pi)=-pi and at x=0 we have xcos(x)=0 since -pi is below the x axis the integral will be negative, is this correct?
 December 13th, 2017, 11:40 AM #4 Member   Joined: Jan 2016 From: Blackpool Posts: 98 Thanks: 2 I also looked at a graph and this makes it clear why it is negative. However, I can't put down in words the answer as it is still not clear to me from the definition of integration. Last edited by skipjack; December 13th, 2017 at 12:12 PM.
 December 13th, 2017, 02:32 PM #5 Global Moderator   Joined: Dec 2006 Posts: 19,718 Thanks: 1806 As $\displaystyle \int_{\pi/2}^\pi x\cos(x)dx = -\int_{\pi/2}^0 (\pi - x)\cos(\pi - x)dx = \int_0^{\pi/2} (x - \pi)\cos(x)dx$, \displaystyle \begin{align*}\int_0^\pi\! x\cos(x)dx &= \int_0^{\pi/2} x\cos(x)dx + \int_{\pi/2}^\pi x\cos(x)dx \\ &= \int_0^{\pi/2} x\cos(x)dx + \int_0^{\pi/2} (x - \pi)\cos(x)dx \\ &= \int_0^{\pi/2} (2x - \pi)\cos(x)dx \end{align*} whose integrand is negative in the interval $(0,\, \pi/2)$.
 December 13th, 2017, 06:53 PM #6 Senior Member   Joined: Sep 2016 From: USA Posts: 475 Thanks: 262 Math Focus: Dynamical systems, analytic function theory, numerics A more general approach which doesn't require exploiting any properties of $\cos x$ is as follows. 1. $\cos x$ is non-negative on $[0,\frac{\pi}{2}]$ and $-\cos x$ is non-negative on $[\frac{\pi}{2},\pi]$. Thus, you have the decomposition into positive integrals $\int_0^\pi x \cos x \ ds = \int_0^{\frac{\pi}{2}} x \cos x \ dx - \int_\pi^{\frac{\pi}{2}} x \cos x \ dx$ Thus, it suffices to show that the first integral is smaller than the second. 2. $\cos x \leq 1$ holds for all $x \in [0,\frac{\pi}{2}]$, so we have the bound $\int_0^{\frac{\pi}{2}} x \cos x \ dx \leq \int_0^{\frac{\pi}{2}} x \ dx = \frac{\pi^2}{8}$ 3. $x \leq \frac{\pi}{2}$ holds for all $x \in [\frac{\pi}{2}, \pi]$, so we have the bound on the second integral $-\int_\pi^{\frac{\pi}{2}} x \cos x \ dx \leq \frac{\pi}{2} \int_{\pi}^{\frac{\pi}{2}} \cos x \ dx = -\frac{\pi}{2}$ 4. Taking these bounds together, we have $\int_0^\pi x \cos x \ dx \leq \frac{\pi^2}{8} - \frac{\pi}{2} = \frac{\pi}{2}(\frac{\pi}{4} - 1) < 0$. The main point of this approach is that it is completely general and illustrates a common "trick" in analysis which is to bound integrals of products by bounding their factors. Often, one breaks the domain into regions of monotonicity like this and the factor which must be bounded changes. Thanks from Jaket1
December 13th, 2017, 08:14 PM   #7
Global Moderator

Joined: Oct 2008
From: London, Ontario, Canada - The Forest City

Posts: 7,879
Thanks: 1087

Math Focus: Elementary mathematics and beyond
Quote:
 Originally Posted by SDK A more general approach which doesn't require exploiting any properties of $\cos x$ is as follows. 1. $\cos x$ is non-negative on $[0,\frac{\pi}{2}]$ and $-\cos x$ is non-negative on $[\frac{\pi}{2},\pi]$.

December 14th, 2017, 05:56 PM   #8
Senior Member

Joined: Sep 2016
From: USA

Posts: 475
Thanks: 262

Math Focus: Dynamical systems, analytic function theory, numerics
Quote:
 Originally Posted by greg1313 Aren't those two statements contradictory?
Please elaborate? This is a basic fact about the $\cos$ function but the method of splitting into positive/negative integrals works in general. I don't know what contradiction you are referring to.

 December 14th, 2017, 06:27 PM #9 Global Moderator     Joined: Oct 2008 From: London, Ontario, Canada - The Forest City Posts: 7,879 Thanks: 1087 Math Focus: Elementary mathematics and beyond "A more general approach which doesn't require exploiting any properties of cos x is as follows." Are the following not properties of the cosine function? "1. cos x is non-negative on [0,π/2] and −cosx is non-negative on [π/2,π]." Nitpicking perhaps, but there it is.

 Tags determine, integral, integrating, sign

 Thread Tools Display Modes Linear Mode

 Similar Threads Thread Thread Starter Forum Replies Last Post Shariq Faraz Calculus 6 October 24th, 2017 07:52 AM zollen Calculus 2 May 8th, 2017 05:22 PM woo Elementary Math 4 June 24th, 2015 12:53 AM chenmq1990 Real Analysis 0 August 7th, 2010 07:29 PM martymon Applied Math 1 June 4th, 2009 06:56 AM

 Contact - Home - Forums - Cryptocurrency Forum - Top