h*********o
发帖数: 151 | 1 If I sent out N1 advertisement version1 with a total cost C1 and n1 people
responded. Then I sent out a different ad, version2, with a drop amount N2
with total cost C2 and n2 people responded.
So the cost per response for the two versions are C1/n1 and C2/n2. Given
confidence level 90%, which test statistic can I use to test if the two
costs per response are significantly different? |
|
