Monetizing Your Data by Andrew Roman Wells & Kathy Williams Chiang

Author:Andrew Roman Wells & Kathy Williams Chiang
Language: eng
Format: epub
ISBN: 9781119356257
Publisher: Wiley
Published: 2017-03-27T00:00:00+00:00

Figure 9.1 Customer Attrition Rate (Annualized)

Correlation Analysis

Another key technique to assist you with determining the relevance of a metric is correlation analysis. A correlation is an association between two or more metrics where a linear dependency or relationship exists, but correlation does not necessarily mean causality. An actionable relationship is the key determinant for our purposes.

What does this mean? Correlation is determining what things happened together. Causality is trying to determine why something happened. A great example comes from an article by David Ritter, “When to Act on a Correlation, and When Not To.” He contemplates the importance of correlation and causality when determining whether to take action based on risk and confidence in the relationship. In this example, the NYC Health department monitors the city for violations and has deployed sewer sensors to collect readings.

These sensors detect the amount of grease flowing into the sewer system at various locations throughout the city. If the data collected shows a concentration of grease at an unexpected location—perhaps due to an unlicensed restaurant—officials will send a car out to determine the source.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.