Re: There are two BOBS.
- From: "\"Luis A. Afonso\"" <licas_@xxxxxxxxxxx>
- Date: Wed, 07 Sep 2005 16:26:56 EDT
concerning your 1:18 PM post
Jerry said:
***That's reasonably clear. One could be picky and argue that it doesn't rule out some kind of systematic method of drawing cards that leads to a predetermined result. For example, the deck could be well-shuffled and you could draw and replace the top card 20 times without violating your description. Note that saying "at random" eliminates the problem entirely. (Also, see my comments below.)***
My response
I do not understand how shuffling the pack (with the later card included) you can have the same card (ace of clubs, for example, twenty times!). You are a true believer.
All the cards are drawn, as usual, from the top, (or from the back) of course. Therefore the drawing is random, the actual position in the pack is SYSTEMATIC!
Jerry said
***There certainly is a random variable W such as you describe. However, it's not clear that this is the variable you're asking about. The phrase "maximum difference...that [is] expected" is ambiguous. "maximum" and "expected" don't work well together when there is only one realization of the random variable. If you had left out "maximum", you would have been in much better shape***.
My response
It lacks the condition that characterizes H0: (I suppose superfluous because assumed). Is it better
____maximum difference that is likely to expected if the pack was al the times thoroughly shuffled ?
..
You say
***Perhaps, you're looking for something like, "20 cards are drawn from the top of a deck of cards. What values of |R-B| suggest that the deck was not well-shuffled?***
My response
Exactly__________perfectly right! (Sorry by do not be careful).
***I don't see any hypothesis test in what you quoted at the top of the post. I see what is likely a question about E(|R-B|).***
The Hypotheses?
It is that the pack was always well-shuffled.(H0)
I posted:
*For example in a real *game* you got less than 6 of one colour you can be sure (at a 5% confidence level) that you did not shuffled thoroughly the deck. If you chose 1% you should tolerate even 5.*
What I intended to say?
If I get a difference less than 6 I can state (at 5% confidence level) that there is not sufficient strong evidence that H0 is untrue.(What else could be?).
(some thing to less than 5 if I chose 1%)
___________licas
.
- Follow-Ups:
- Re: There are two BOBS.
- From: Jerry Dallal
- Re: There are two BOBS.
- References:
- There are two BOBS.
- From: \"Luis A. Afonso\"
- There are two BOBS.
- Prev by Date: Re: Geometric average vs Arithmetic average
- Next by Date: Re: Uncertainty across a boundary - Restated!
- Previous by thread: Re: There are two BOBS.
- Next by thread: Re: There are two BOBS.
- Index(es):
Relevant Pages
|
