AnalyticBridge

Social Network For Analytic Professionals

The TEST & CONTROL strategy is a popular strategy of the Marketing world used to compare sales across different schemes to measure the effectiveness of each individual scheme/campaign.

I've got about 100 regions, which I'll need to split into 3 groups of similar characteristics in terms of sales, redemptions & transactions along with other factors if any, in the region.

What is the way by which I can make sure that I have at hand, three groups which are very similar in characteristics? A random sampling will not mostly work, since there is no specification of similarity of measures. Is there an alternative?

Share Twitter

Reply to This

Featured


Advertisement

© 2010   Created by Vincent Granville

Badges  |  Report an Issue  |  Privacy  |  Terms of Service