In this issue, I am borrowing from my recent keynote at Predictive Analytics World on “the 5 Myths of Predictive Analytics” with this SAS interview and a story board.
Predictive Analytics is used by various organizations to analyze current and historical facts to make predictions about health care, financial collection activities, customer behavior, and customer retention. Although Predictive Analytics is a powerful optimization technique, it is often misunderstood, and thus misused.
The top 5 myths of Predictive Analytics, I am going to address today are:
- Predictive Analytics is new: In the recent history, the first credit scoring model was built in 1930′s by Fischer and Durand, but predictive modelling technique goes even further back thousands of years – the use of Indian astrological chart in arranged marriages being one such example.
- Perfect Prediction:Often while building the model, it is clear to all that model prediction has a probability associated with it, but upon successful use, there is often a misplaced sense of perfectness in the scores.
The next 3 myths below with fun illustration by Matter Solutions.
- Good Tool = Good Model: With tremendous development on the tool front with better GUI as well as higher automation, people new to this field often mistakenly believe, that good model can be built by pressing the “build regression model” button automatically.
- Good Model = Good $: This is one of those highly prevalent myths that even experienced analyst fall for, often finding them frustrated that nobody in the business seems to care for the amazing model they have built. Good model generates the business impact, only when right stakeholders are brought into the analytics process at the right time building proper alignment toward actionability.
- Build and Forget: Models gets stale over time and if not maintained, often stops delivering the incremental it started with.