[R] Numerical question

[Duncan Murdoch]

On 25/07/2008 5:06 PM, Zhang Yanwei - Princeton-MRAm wrote:
> Hi all,
>    I have n independent variables A_1, A_2, A_3,......,A_n, and each with known variances var(A_1), var(A_2),..., but unknown mean. How can I get the approximation of the variance of the product of the variables using numerical computation, i.e. var(A_1*A_2*A_3*.....*A_n)? Thanks.

You can't, in any reasonable sense.  The variance depends strongly on 
the means, which you say you don't have.

If you want to be unreasonable, then the answer is that the variance is 
approximately 1.

Duncan Murdoch