Re: Two-envelope problem

[Tuan V Nguyen <t.nguyen@garvan.unsw.edu.au>]

Hi Vu~,

        Good argument! It is good to see you in the game. I have thought
long and hard about this problem since my friend gave me last week. I will
put my view on anh Du~ng's solution when I come home. In general, I feel
that the concept of expectation (as used by anh Du~ng) may not be
appropriate. The concept of expectation, as features prominently in
statistics, is based on the assumption that we can repeat the "experiment"
infinite number of times. You may find 100 first, OK. What about you find
$50 first, and so on. I will put my view on this soon ...

        Tuan

At 01:06 AM 4/18/97 -0500, you wrote:
>I would disagree with this solution. This problem is a classical paradox
>in statistical decision theory, often being stated to fool novices.  The
>punchline is that if you follow this line of reasoning (meaning willing
>to exchange the envelopes), then you would deem it wise to exchange the
>envelopes one more time and indeed, exchange the envelopes infinitely
>many times, an apparent antinomy!
>
>More later,
>Vu~
>
>Dung Trong Nguyen wrote:
>
>  1) The expectation value of keeping the $100
>
>          1 x $100 = $100
>
>  2) The expectation value of choosing the other
>
>          1/2 x $50 + 1/2 x $200 = $125
>
>  So it is better to take the risk.
>
>  d~
>