# How to fine the required sample size

1. Feb 5, 2012

### alias

1. The problem statement, all variables and given/known data
Given info: cost1=$4, σ1=10, W1=N1/N=0.4 cost2=$9, σ2=20, W2=N2/N=0.6

The second part: Find the sample size required, under this optimal allocation to make
V(ybar) = 1. Ignore the finite population correction factor.

2. Relevant equations
The equation I used for the first part is nh = n[(Nh)(σh^2)/(ch)]/[(summmation j=1 to H) ((Nj) (σj^2)/(cj))] , where H = 1, 2.

3. The attempt at a solution
The first part of this question asks for the values of n1/n and n2/n that minimize total cost for a given value of the variance, V(ybar). It is a stratified sample.
My answer is n1/n = 0.1818 and n2/n = 0.8181

Can anyone help me with the second part? I'm very lost, any ideas would be appreciated. Thanks

Last edited: Feb 5, 2012