home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: rec.gambling
- Path: sparky!uunet!walter!news
- From: Jonathan Rosenberg <jxr@thumper.bellcore.com>
- Subject: Re: Not the Monty Haul problem
- Message-ID: <1992Nov21.162331.12407@walter.bellcore.com>
- Sender: news@walter.bellcore.com
- Nntp-Posting-Host: bambam.bellcore.com
- Organization: Bellcore
- Date: Sat, 21 Nov 92 16:23:31 GMT
- Lines: 85
-
- It's true that this discussion really belongs on "something".math, but
- since it was asked here & I've seen two incorrect answers, I'm going to
- provide an answer here. This problem is actually much tricker than it
- seems.
-
- Jeff Tang Asks:
-
- > Question:
-
- > You're given two envelopes and told that the contents are determined
- > randomly, except that one has twice as much money as the other. You open
- > one envelope and find 1 unit. Now, you're given the opportunity to either
- > take the unit you found, or give it up and take the contents of the other
- > envelope. What do you do?
-
- > Answer 1:
- > Well, since there wasn't really any information revealed by opening the
- > envelope, switching to the other envelope wouldn't be any different than
- > choosing that envelope in the first place. Therefore, it doesn't matter if
- > you switch.
-
- > Answer 2:
- > Well, the other envelope has either .5 or 2 units. Since there's a 50%
- > chance of each of these, you'll come out with 1.25 units on average as
- > opposed to sticking with the single unit, so you should switch.
-
- > What's wrong with one (or both) of these answers?
-
- Believe it or not, BOTH answers are wrong, though Answer #2 is actually
- closer to a correct answer. No, this is not a trick. Let me explain.
-
- The difficulty is that the problem is not completely specified &, so, is
- unanswerable. Let's look at the problem statement again:
-
- You're given two envelopes and told that the contents are determined
- randomly, except that one has twice as much money as the other.
-
- The difficulty lies in the phrase "contents are determined randomly".
- We can assume without loss of generality that the contents are
- determined as follows: choose a number U & place that many units in an
- envelope. Now place 0.5U or 2U (determined by a flip of a coin, say)
- units in the other envelope.
-
- To answer the question correctly requires knowing the distribution of U
- (i.e., what is the distribution of the random variable determining the
- contents). Without knowing the distribution, there is no answer.
-
- You're probably thinking that this is nit-picking & that it's obvious
- what the distribution is. Well, I can think of 2 obvious possibilties:
-
- 1) U is chosen with a uniform distribution from the interval [0, M].
- That is, the minimum that can be put in an envelpe is 0 units & the max
- is M units. In this case, there is a simple algorithm for maximizing
- the expected value:
-
- If the enveople you choose has <= M/2 units in it, switch to the other
- envelope.
- Otherwise, stick with you first choice.
-
- Notice that answer #2 is a variation of this for the case M=2.
-
- 2) U is chosen with a uniform distribution from the interval [0,
- infinity). That is, the minimum number of units is 0, but there is no
- maximum. I suspect that this is implicitly what everyone assumed the
- distribution was.
-
- If this is the correct meaning, then it's easy to deal with. That's
- because the construction specified is impossible: it's not possible to
- choose a number uniformly from an infinite sequence. Thus, the problem
- is ill-defined & there is no answer.
-
- ------
-
- On rethinking this, I realized that you might argue that the problem is
- well-defined & that the chooser just is not told the distribution used.
- That seems to be ok. In that case, answer #1 is correct. Answer #2 has
- flawed logic because it is calculating an average of repeated trails
- that are not like the original trial. I.e., answer #2 assumes that in
- each trail the envelopes will contain either 0.5 & 1 units or 1 & 2
- units. But, that ain't the original problem. The original problem says
- that you have no information about the amounts in the envelopes (other
- than 1 is twice the other).
-
- JR
-
-