This assignment needs to be completed in SPSS… For each question, there needs to be screen shots for each step of each question with completed questions. Second, previously I’ve gotten many wrong answers… please I need correct answers to the questions!

The purpose of this assignment is to perform neural nets
classification, interpret the results, and analyze whether or not the
information generated can be used to address a specific business
problem.

For this assignment, you will use the “Credit Card Defaults
Oversampled” data set from the Topic Materials. Note that this data set
includes the same information as the Topic 5 assignment data set,
however, the default cases have been randomly oversampled to balance the
defaulters vs. the nondefaulters. This is one method to increase the
accuracy of a rare outcome (i.e., default).

You are an analyst for a credit card company. Management wants to
know if there are any early signs that indicate whether customers will
default on their credit cards. If these indicators can be identified,
then more scrutiny can be placed on customer transactions in an effort
to avoid losses. For this project, management is interested in very
accurate predictions. Management would like to receive a list of ranked
variables related to default, however, understanding the specific
relationships between the predictors and the likelihood of default is
secondary.

Question 1: Partition the data to create a training data set (70%) and test data set (30%).

Question 2: Build a neural network using
the Multilayer Perceptron method with the training data and “Default” as
the target. Make sure to output predictor importance. Include the
“Default Neural Network” output when submitting the answer.

1. What is the ranking of the variables? Include the “Predictor Importance” output when submitting the answer.
2. What is the accuracy of the model when using the training and test
   data? Include the “Misclassification Table” outputs when submitting the
   answer.

Question 3: Rerun the model but only use the top six variables.

1. What is the accuracy of the model when using the training and test
   data? Include the “Misclassification Table” outputs when submitting the
   answer.
2. Did the accuracy improve? Explain why.
3. Produce a ROC curve when using the training and test data and
   interpret the results of the ROC curves. What do the curve and the
   45-degree line represent?

Question 4: Explain how the selected model
can be used to solve the business problem. What are the advantages and
disadvantages when building a neural net model vs. the simpler
classification model like the one completed in the Topic 5 assignment?

General Requirements:

Submit the answers to Questions 1-4 as a Word document.