Results 1 to 2 of 2

Math Help - Variance of the sum of two poisson random variables

  1. #1
    Junior Member
    Joined
    Dec 2010
    Posts
    62

    Variance of the sum of two poisson random variables

    Hi, I have this problem, the solution of which is unclear to me. It's as follows.

    Let the Poisson random variable U be the number of calls for technicals assistance received by a computer firm during the firm's normal 9 hour day. Suppose the average number of calls per hour is 7.0 and that each call costs the company $50. Let V be a Poisson random variable representing the number of calls for assistance received during the remaining 15 hours of the day. Suppose the average number of calls per hour for this time period is 4.0 and that each call costs the firm $60. Find the expected cost and the variance of the cost associated with calls received during a 24-hour day.

    Now, the expectation and the variance for a Poisson random variable are the same, right? And as these are independent random variable the expected cost of the sum should be the sum of the individual expectations. i.e.

    \mathrm{E}[U + V] = \mathrm{E}[U] + \mathrm{E}[V] = \$3150 + \$3600 = \$6750

    This squares with the solution manual. However, it gives the variance as $373500.

    But if these are independent random variables, why isn't the variance the sum of the individual variances, and in this case the same as the expectation? i.e. $6750

    This has got to be a typo right? Or am I missing something. My experience with presumed typos in other maths books is it's usually the latter ;-) so I thought I'd check with you cats, MD.

    p.s. shouldn't the units of the variance be \$^{2} too?
    Last edited by Mathsdog; June 11th 2012 at 10:50 PM.
    Follow Math Help Forum on Facebook and Google+

  2. #2
    mfb
    mfb is offline
    Junior Member
    Joined
    Jun 2012
    From
    Germany
    Posts
    72
    Thanks
    5

    Re: Variance of the sum of two poisson random variables

    The expectation and the variance for a Poisson random variable are the same. But the poisson-distributed random variables here are the number of calls, not the cost. Both the expectation and variance for the number of calls are dimensionless.
    If you multiply a random variable with a constant factor, the variance gets scaled with the square of this factor. Therefore, if u is the number of calls during daytime and U the costs, Var(U)=(50$)^2 * var(u) = (50$)^2 * E(u) = (50$) * E(U) = 157500$^2
    Similarly, Var(V)=60$*E(V)=216000$^2 and Var(U+V)=373500$^2.
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. Variance, Means and Random Variables.
    Posted in the Statistics Forum
    Replies: 2
    Last Post: January 31st 2011, 10:58 AM
  2. Independent Random Variables - Variance of 2 Die
    Posted in the Discrete Math Forum
    Replies: 1
    Last Post: November 29th 2010, 09:54 PM
  3. Marginal variance with two random variables
    Posted in the Advanced Statistics Forum
    Replies: 2
    Last Post: October 20th 2010, 05:43 AM
  4. What is the variance of the division of 2 random variables?
    Posted in the Advanced Statistics Forum
    Replies: 4
    Last Post: March 23rd 2010, 12:23 PM
  5. random variables and variance
    Posted in the Advanced Statistics Forum
    Replies: 2
    Last Post: November 26th 2008, 01:25 AM

Search Tags


/mathhelpforum @mathhelpforum