Aggregation and Further Analytics

At this point, I’d like to consider the analytics that we just performed. We did these analyses under some assumptions. One of the most basic assumptions is that the variables are statistically significant in our predictive regression model for the binary outcome response of customer risk of revenue decline. In the case where we sampled customers from our customer base, this type of analysis is known as ad hoc or non-experimental research study designs. This means that in the study we just performed we didn’t have complete control of all of the designed experimental input factors. If we designed the study, we could have chosen certain control and test groups, or confined the study to a certain region of customers, ...

Get Strategic Analytics and SAS now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.