Reference no: EM132970310

Reference : chapters 4 & 5 of Siegel's book.

[1] Siegel, E. (2016). Predictive analytics: The power to predict who will click, buy, lie, or die (Revised and Updated ed.).

Note: Firstly provide a summary in detail and answer the below questions in two paras each.

After giving the summary, answer the following questions.

a. What is the difference between a univariate and a multivariate predictive model? What is an uber-variable?

b. What is a decision tree? Use an example to show root, leaf and segment in a tree. Write down one of the patterns in the tree.

c. What is the goal of the data preparation step? What should the resulting training data look like?

d. Write down the business rule for the rightmost path (if answering no at every step) in the Chase mortgage decision tree on page 163.

e. Explain the meaning of 'the model has a lift of 3.0 at the 20% mark.'? How does such a model lower churn?