SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Politics : Formerly About Advanced Micro Devices

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: andreas_wonisch who wrote (128122)11/10/2000 6:43:25 PM
From: Ali Chen  Read Replies (1) of 1583744
 
Andreas, OT<The expectation is evidently n*p = 2000 * 0.5 = 1000 and the probability for getting at least 1000 heads is 0.5. That's also what intuition is telling us...>

Your formula n*p and is absolutely correct, but
the formal conclusion about 0.5 is not. This is
exactly the basic definition of 0.5 probability:
when flipping a coin 2000 times, expect any side
to appear about 1000 times, almost certainly.

<Or do you expect to get with almost certainty (98.5%) at least 10 heads when flipping a coin 20 times?>

With 20 attempts I expect to have 10 heads with
probability of 1-0.5/sqrt(20), or 94%. As the
number of attempts grows, the confidence interval
shrinks as square root from the total number of
attempts, possibly with some numerical factor
of the order of 1 I don't remember exactly.
With infinite amount of attempts you will get
1000 heads for sure.

Disclaimer: the terms and definitions used above
may be slightly different from one science field
to another.

Regards,
- Ali

P.S. I just saw your precise Math. The width of
your distribution function narrows as sqrt(N) too.
We are talking in different terms here. To get _exactly_
1000 out of an _event_ with 2000 attempts, you may be
right. But if we think about getting 999, 1000, or 1001
in a single 2000-fold attempt, the probability is 0.98.
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext