## Forum Posts

zzz aaa

Jun 21, 2022

In Business Forum

I shared the basic principles of ABtest before. Some friends asked: If I have not only two versions of AB, but three versions of Whatsapp Database ABC for testing, can I still use the ABtest method? Of course it works! It's just that the statistical method used is changed to: Whatsapp Database analysis of variance, and I will simply share it with you today. 1. The difference between multi-version and AB version If there are only two versions of AB to compare, then when doing hypothesis testing, the null hypothesis is: the mean/proportion of version A = the mean/proportion of version B. After that, as long as we can use the Whatsapp Database test results to overturn the null hypothesis, we can show that the AB versions have different means/proportions, and thus argue which version is better. The corresponding statistics are also designed accordingly. Whatsapp Database The problem becomes a little more complicated when there are multiple versions of ABCDE... . Because it is very likely that in this pile of versions, some have differences and some do not. If you test one by one, you have to compare n times, which is time-consuming and labor-intensive. How to complete the test efficiently? Second, the basic idea of multi-version testing A two-step method can be used. Step 1: Do an analysis of Whatsapp Database variance to test whether the means of these versions are equal. At this point, the original hypothesis is: A version = B version = C version = D version... In short, assume that they are all the Whatsapp Database same. If the assumption is not overturned, it means that everyone has the same mean, and there is no need to do the second step at all. If it can be overturned, then look at the gap between the two.

0

0

1

# zzz aaa

More actions