Thank you Professor Neal,

I knew it had something to do with indicators but I couldn't really
consolidate it into a solution.


[EMAIL PROTECTED] (Radford Neal) wrote in message news:<[EMAIL PROTECTED]>...
> >[EMAIL PROTECTED] (Xiao Li) wrote in message 
> >news:<[EMAIL PROTECTED]>...
> >
> >> A theater has a set of seats consisting of four rows and twelve
> >> columns for a total of 48 seats.  10 people walk into the theater and
> >> each person takes a seat at random.  Let X be the number of people
> >> with someone sitting directly in front of them.  What is the expected
> >> value of X?
> 
> In article <[EMAIL PROTECTED]>,
> Ray Koopman <[EMAIL PROTECTED]> wrote:
> 
> >E(X) = 12*E(Y), where Y is the number of people in any given column
> >that have someone sitting directly in front of them...
> 
> Another, simpler, way is to note that E(X) = sum_i E(Y_i), where Y_i
> is the number of people sitting directly in front of person i (in 
> some arbitrary labelling).  Note that Y_i is either 0 or 1, so 
> E(Y_i) = P(Y_i=1).  Note also that all the Y_i are the same (nothing
> distinguishes one person from another a priori).  So E(X) = 10 * E(Y_1).
> 
> If person i sits in the front row, no one will be in front of them.
> Otherwise, as happens with probability 3/4, the probability that someone 
> will be directly in front of them is just the probability that one of the 
> other 9 people will sit in the chair in front of them.  These other 9
> people will occupy 9 of the 47 other chairs.  The probability that the 
> chair in front is one of the occupied chairs is therefore 9/47.
> 
> The final answer is therefore E(X) = 10 * (3/4) * (9/47) = 1.43617.
> This matches the answer obtained by Ray Koopman.
> 
> ----------------------------------------------------------------------------
> Radford M. Neal                                       [EMAIL PROTECTED]
> Dept. of Statistics and Dept. of Computer Science [EMAIL PROTECTED]
> University of Toronto                     http://www.cs.utoronto.ca/~radford
> ----------------------------------------------------------------------------
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to