Data Mining and Predictive Analytics for Business Decisions by Andres Fortino

Author:Andres Fortino
Language: eng
Format: epub
Publisher: Mercury Learning and Information

FIGURE 7.29 Correlation between gross approval amount (signified by variable GrossApproval) and firm size (signified by variable JobsSupported).

Run the Linear Regression function, and create a table for coefficients to get the result shown in Figure 7.30.

FIGURE 7.30 Linear regression between between gross approval amount (signified by variable GrossApproval) and firm size (signified by variable JobsSupported).

CASE STUDY 7.3: LOGISTIC REGRESSION USING THE SFO SURVEY DATA SET

Continuing with our study of the overall ratings the SFO passengers give the airport, let us approach it from a binary point of view, which can be analyzed using logistic regression. We can bin the Q7ALL variable into two categories. This first is “Do you like this airport?,” where “Yes” = 5, 4, and “No” = 3, 2, 1; we replace the 6 and 0 scores with blanks. Likewise, we can bin the Net Promoter Score, NETPRO, into two categories: “Would you recommend this airport to your friends?,” where for “Yes,” NETPRO= 10, 9, 8, and for “No,” NETPRO = all other numbers; and we leave a blank for any other response. We will predict their response based on their other responses to other airport features (such as food and parking) and the other Q7 answers as features or input variables.

The question we will answer using logistic regression for this case is

Can we predict what factors are most important for customers to give us a good score?

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.