Must use JMP PRO !4! Marketers are interested in understanding what factors determine the likelihood of attracting new customers (predicting customer acquisition), what factors determine the next best offer to make to a customer to elicit purchase (next best offer) and what factors determine customer churn (leaving).
These are 3 situations in which marketers use predictive analytics. In this exercise, we use logistical regression to predict a binomial dependent variable (for example, 1. churn /0. No churn) to predict customer response. Part 1 requires that you replicate an example from the text and part 2 requires that you develop a model that predicts customer churn. Objective: To develop familiarity with JMP and doing predictive analytics.
JMP PRO 14 Part 1: Read Chapter 5 of Creating Better Models with JMP PRO (pages 101 to 114 – stop at Titanic Passengers). Use the JMP PRO Application and the data table or data file lostsales Download lostsales(Download the file and open it in JMP PRO) to replicate the following analysis from chapter 5. You are replicating the lost sales example which involves predicting "order status" using 3 variables, namely quote, time to delivery and part type. Chart/graph each variable individually.
To save your output use the Insert/clippings function in MS Word to cut and paste your results into a Word document (make sure its readable). Do the “Fit Y by X” analysis using status as Y and the remaining variables as X variables. Estimate a logistic regression to predict order status (Y) using the 3 predictor variables. The Y variable is categorical (order/no order) so you will select nominal logistic in the personality window.
Review the parameter estimates window and identify and remove the insignificant variable (p< or=.05) and rerun the model. Save the output results for your report. Select fit details and report on the misclassification rate of the model. Pay attention to the coefficients and the signs (negative or positive) for each variable in interpreting your results. Select the red arrow at the top of the results page at Nominal Logistics Fit and scroll down to save probability formula and select.
Then view the data table. There are now 4 new columns in your data. The right most columns is what the model predicts based on the data. If you compare the predictions with the actual you can identify the cases where the model makes incorrect predictions. Include a screen shot of the data table with the new results. Take note that if you add new data for the predictor variables to the spreadsheet, the model will predict the dependent variable: churn or lost sales
Part 2: Predicting Customer Churn Customer retention is a big challenge in the ultracompetitive mobile phone industry. A mobile phone company is studying factors that predict customer churn, a term used for customers who have moved to another service provider. The Task The company would like to build a model to predict which customers are most likely to move their service to a competitor.
This knowledge will be used to identify customers for targeted interventions, with the ultimate goal of reducing churn. Use the data file churn2-bbm.jmp Download churn2-bbm.jmpin the files menu The sample data set consists of 3,332 customer records. The response variable of interest is the column called Churn, which takes two values: True The customer has moved to another service provider. False The customer still uses “our” service.
The potential predictors are primarily related to service use and account. In the model specification window select true for Target Level to indicate that your model is predicting the likelihood that customers churn (inverse of the retention rate). Deliverables: Note that the churn example is used in the neural network predictive modeling exercise in chapter 7. I don’t require you to do the neural network model. I only required that you use the data to estimate a logistic regression model predicting the likelihood of customer churn. From this exercise identify the variables that significantly predict the likelihood of a phone customer churning. Please include the results page in your report and a page view of the data table with the predicted results from step 6 above. Note that many of the variables will be insignificant and you will need to delete them and rerun the model. You can capture this information using a screen clip in Word. Considering the variables that predict customer churn, identify 4 bulleted (each explained in no more than 3 sentences) recommendation to reduce phone customer churn. Explain in your words the steps you took to develop the model and your assessment of the value or usefulness of the model.
CS 340 Milestone One Guidelines and Rubric Overview: For this assignment, you will implement the fundamental operations of create, read, update,
Retail Transaction Programming Project Project Requirements: Develop a program to emulate a purchase transaction at a retail store. This
7COM1028 Secure Systems Programming Referral Coursework: Secure
Create a GUI program that:Accepts the following from a user:Item NameItem QuantityItem PriceAllows the user to create a file to store the sales receip
CS 340 Final Project Guidelines and Rubric Overview The final project will encompass developing a web service using a software stack and impleme
