Inspired by Geoff Pilling

In Timbuktu, each woman is equally likely to have 0, 1, or 2 children.

If Ella has more children than Bella, what is the probability that Ella has more children than Frabduzella?


Inspiration

1 3 \frac{1}{3} 4 9 \frac{4}{9} 5 9 \frac{5}{9} 2 3 \frac{2}{3}

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.

3 solutions

There are 9 equally likely instances ( E , B , F ) (E,B,F) , (where E , B , F E,B,F represent the respective number of children of Ella, Bella and Frabduzella), in which E > B E \gt B , namely

( 2 , 0 , 0 ) , ( 2 , 0 , 1 ) , ( 2 , 0 , 2 ) , ( 2 , 1 , 0 ) , ( 2 , 1 , 1 ) , ( 2 , 1 , 2 ) , ( 1 , 0 , 0 ) , ( 1 , 0 , 1 ) , ( 1 , 0 , 2 ) (2,0,0), (2,0,1), (2,0,2), (2,1,0), (2,1,1), (2,1,2), (1,0,0), (1,0,1), (1,0,2) .

In 5 of these cases E > F E \gt F as well, so the desired probability is 5 9 \boxed{\dfrac{5}{9}} .

In general, if each woman is equally likely to have 0 , 1 , 2 , . . . , n 0,1,2,...,n children and E > B E \gt B , then the probability that E > F E \gt F as well is

k = 0 n k 2 ( n + 1 ) k = 0 n k = n ( n + 1 ) ( 2 n + 1 ) 6 n ( n + 1 ) 2 2 = 2 n + 1 3 ( n + 1 ) \displaystyle \dfrac{\sum_{k=0}^{n} k^{2}}{(n + 1) \sum_{k=0}^{n} k} = \dfrac{\dfrac{n(n + 1)(2n + 1)}{6}}{\dfrac{n(n + 1)^{2}}{2}} = \dfrac{2n + 1}{3(n + 1)} , which goes to 2 3 \dfrac{2}{3} as n n \to \infty , consistent with the continuous-variable inspiration problem.

Zain Majumder
Oct 21, 2018

Since we are given that Ella has more children than Bella there are three possibilities:

  • Ella has 2 2 children and Bella has 1 1 child
  • Ella has 2 2 children and Bella has no children
  • Ella has 1 1 child and Bella has no children

In the first two cases, there is a 2 3 \frac{2}{3} probability that Frabduzella has fewer children than Ella. In the third case, the probability is 1 3 \frac{1}{3} . Therefore, the answer is 2 3 × 2 3 + 1 3 × 1 3 = 5 9 \frac{2}{3} \times \frac{2}{3} + \frac{1}{3} \times \frac{1}{3} = \boxed{\frac{5}{9}} .

Calvin Lin Staff
Oct 17, 2018

[This is not a solution.]

In Geoff's problem, the answer is 2 3 \frac{2}{3} .
What is the answer different in this problem?
What happens if the woman are equally likely to have 0, 1, 2, and 3 children?

As you've pointed out in Geoff's problem, the difference is that it's said that the 3 women are of different heights, but here it's not said that they have different number of children.

7 12 \dfrac{7}{12} final count in the case of 0, 1, 2, 3 children. Okay, finally. Should have done this by computer. This agrees with Brian's results.

Michael Mendrin - 2 years, 7 months ago

Log in to reply

Edit: Was the 13 24 \frac{13}{24} in reference to "0,1,2, and 3 children"?
If it was in reference to "0, 1, and 2 children", then I was debating adding in "strictly more" and it seems like I should do so?

Calvin Lin Staff - 2 years, 7 months ago

Log in to reply

Yes, "0,1,2, and 3 children", and, yes, "strictly more". Let me go count them up again. I did it in an hurry, but now I think I have time.

Edit: Never mind the 13 24 \frac{13}{24} . See above.

Michael Mendrin - 2 years, 7 months ago

I'm getting 14/24 = 7/12 for the 3 children case. Just working on a general formula for n children....

Edit: I get a general formula of 2 n + 1 3 ( n + 1 ) \dfrac{2n + 1}{3(n + 1)} , which goes to 2 3 \dfrac{2}{3} as n n \to \infty , consistent with Geoff's continuous-variable problem.

Brian Charlesworth - 2 years, 7 months ago

Log in to reply

Indeed.

With reference to the comment in the inspiration

If the heights are distinct, the probability is 2/3.
However, if there was a non-zero probability that 2 girls could be of the exact same height, then the answer could be different. IE In the case where the heights are x > y = z x > y = z , the probability is 1, and in the case where the heights are x = y > z x = y > z , the probability is 0.

Thus, the total probability is n ( n 1 ) ( n 2 ) × 2 3 + n ( n 1 ) × 1 + n ( n 1 ) × 0 n ( n 1 ) × n = 2 n 1 3 n \frac { n(n-1)(n-2) \times \frac{2}{3} + n(n-1) \times 1 + n(n-1) \times 0 } { n(n-1) \times n } = \frac{ 2n-1}{ 3n } for n 0 , 1 n \neq 0, 1 .
(I'm indexing by the number of options, not by the number of children, so our answers agree.)


This also shows how grouping the cases makes the calculation easier than trying to list out all possible cases. This "iterated expectation" calculation is extremely useful in such problems where the "conditional probability" version is much simpler to consider.

Calvin Lin Staff - 2 years, 7 months ago

(As a point of clarification for others)

Technically "3 women are of different height" is not needed in Geoff's problem, because the probability of that happening is 0 (under the implicit assumption that height is a purely continuous distribution).

The true distinction in the problems is that mine is a discrete distribution where the probability that 2 values are the same is non-zero.

Calvin Lin Staff - 2 years, 7 months ago

Log in to reply

Hey Calvin Lin, remember me?

Rui Chang Lu - 1 year, 1 month ago

@Calvin Lin - Sir, this is off-topic on this problem, but can you please post a solution to this problem of yours. I was fascinated by it, but not able to solve it. I asked a little help from other members, but it seemed that the ones that replied didn't know either. I request you to please post a solution. Thank you :)

A Former Brilliant Member - 7 months, 2 weeks ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...