Random Chord Division

Calculus Level 5

A chord is formed by connecting two points that are randomly and uniformly chosen along the circumference of a unit circle, dividing the circle into two regions.

The expected value of the area of the larger region is A π + B π , A\pi + \dfrac{B}{\pi}, where A A and B B are rational numbers . What is A + B ? A+B?


The answer is 1.75.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

1 solution

Relevant wiki: Geometric Probability - Problem Solving

Let the angle that one of the points on the circumference of the circle make at the origin, be θ 1 \theta_1 . Let θ \theta be the angle subtended by the chord joining this point to another independently and uniformly chosen point on the circumference. Given θ 1 \theta_1 , the angle θ \theta , is just the measurement of the angle of the second point on the circumference with respect to θ 1 \theta_1 . But since the second point is chosen independently of the first point, and is chosen uniformly, θ \theta is distributed as U [ 0 , 2 π ) \mathcal{U}[0,2\pi) , independent of θ 1 \theta_1 . The chord subtending angle θ \theta at the center of the circle divides the circle into two regions of areas A 1 = θ sin θ , A 2 = 2 π A 1 A_1=\theta-\sin\theta,\ A_2=2\pi-A_1 . It is easy to check that A 2 A 1 A_2\ge A_1 if θ [ 0 , π ] \theta\in[0,\pi] , and A 1 > A 2 A_1>A_2 , if θ ( π , 2 π ) \theta\in (\pi,2\pi) . Thus the expected area, conditioned on θ 1 \theta_1 is 1 2 π [ 0 π ( 2 π θ + sin θ ) d θ + π 2 π ( θ sin θ ) d θ ] = 1 2 π [ 2 0 π ( 2 π θ + sin θ ) d θ ] = 1 π [ 2 π π π 2 2 + 2 ] = 3 π 4 + 1 π \frac{1}{2\pi}\left[\int_0^{\pi}(2\pi-\theta+\sin\theta)d\theta+\int_{\pi}^{2\pi}(\theta-\sin\theta)d\theta\right]\\=\frac{1}{2\pi}\left[2\int_0^{\pi}(2\pi-\theta+\sin\theta)d\theta\right]\\=\frac{1}{\pi}\left[2\pi\cdot \pi-\frac{\pi^2}{2}+2\right]=\frac{3\pi}{4}+\frac{1}{\pi} Since the result is independent of θ 1 \theta_1 , this is also the unconditional expectation, which gives the answer 1.75 \boxed{1.75} .

Great solution. There are a couple of other ways to deal with the casework; e.g., just assume θ [ 0 , π ) \theta \in [0,\pi) via symmetry.

However, there's one point you haven't justified fully. Why is picking two points uniform on the circumference equivalent to picking an angle uniformly (as opposed to some other distribution)?

Eli Ross Staff - 4 years, 7 months ago

Log in to reply

I have revised my answer to include a justification of the point raised.

Samrat Mukhopadhyay - 4 years, 7 months ago

Log in to reply

Great, thanks!

I know it may seem obvious that angles correspond directly to arc length here, but this point (mapping a uniform distribution to another distribution) is often overlooked and can lead to pitfalls in similar problems.

Eli Ross Staff - 4 years, 7 months ago

Log in to reply

@Eli Ross Yes, apparently simple things can be extremely deceiving in many cases.

Samrat Mukhopadhyay - 4 years, 7 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...