3 comments

[ 3.5 ms ] story [ 16.7 ms ] thread
Read the article. Not sure I understand how this is used in real world applications? Is the idea that the sequential analysis method returns an alert / flag once there is sufficient data to conclude that an A or B has been selected?
Yep, the method will alert you when there is enough data to be able to conclude which hypothesis is true.
A/B testing is probably the best experimental test method to implement for any projects, but it does take a lot of time and investment to complete it. But thank you for explaining this Sequential Generalized Likelihood Ratio test. Definitely is a faster shortcut for experimentation.