Prev: Question on queuing theory
Next: most common algorithms and design patterns ...
From: Tim Little on 5 Dec 2009 21:51
On 2009-12-06, Ben Bacarisse <ben.usenet(a)bsb.me.uk> wrote:
> But the expected value we want is for the ratio #S_k/sum_t(#S_t)

I don't think it has been clarified yet whether the original poster
(whose identity eludes my memory) wanted E(#S_i) / E(sum_t(#S_t)) or
E(#S_i / sum_t(#S_t)). In all probability they have lost interest by
now and we will never know.


- Tim
From: Ben Bacarisse on 5 Dec 2009 22:23
Tim Little <tim(a)little-possums.net> writes:

> On 2009-12-06, Ben Bacarisse <ben.usenet(a)bsb.me.uk> wrote:
>> But the expected value we want is for the ratio #S_k/sum_t(#S_t)
>
> I don't think it has been clarified yet whether the original poster
> (whose identity eludes my memory) wanted E(#S_i) / E(sum_t(#S_t)) or
> E(#S_i / sum_t(#S_t)). In all probability they have lost interest by
> now and we will never know.

No, we probably won't. However, I was responding to your post in
reply to Barb Knox. Unless I've misunderstood, she was discussing the
expected value of the ratio, not the ratio of the expected values. If
you were just re-stating that you read the problem differently, then I
missed your point for which I am apologise. I thought you could not
see how Barb got 1/2 in both cases.

--
Ben.
From: Reinier Post on 9 Dec 2009 14:23
Paul E. Black wrote:

>An analog is the baby gender ratio paradox. Suppose a society values
>boys babies, but not girl babies. Specifically when a couple has a
>girl, they don't have any more children. Assuming equal chance of
>having a girl or boy, what's the percentage in the overall population?
>An incorrect intuition is that one will see families of 3 boys and 1
>girl or 4 boys and 1 girl, so there must be more boys. Actually the
>ratio is exactly 1/2. How could that be? Families with several boys
>will be scarce.

Unfortunately, such societies do exist, and your 'specifically'
doesn't quite hold for them.

--
Reinier
First  |  Prev  | 
Pages: 1 2 3
Prev: Question on queuing theory
Next: most common algorithms and design patterns ...