Home Assignment

AI and ML Part I

For this assignment, you need to work on SAS Via with a data set named PVA. It contains data that represents charitable donations made to an American veterans association. The data represent the results of a mail campaign to solicit donations. The data set contains 106,546 observations and 29 variables the following information:

• a flag to indicate respondents to the appeal (Target Gift Flag)
• the dollar amount of their donations and (Target Gift Amount)
• respondents PVA promotion and giving history
• demographic data of the respondents

First, you will need to create a new report in Viya and add PVA as data source to that project. You may search PVA under Available sources OR click on Data Sources and then find folder SVSO83, where dataset is stored. If you are not sure, watch Demo 7, which shows how to add data source.

Second, you need to change variable Target Gift Flag from Measure to Category type. That variable takes two values 1 if donated, 0 otherwise. (Similar transformation was performed with outcome variable in VS_Bank see Demo 8).

1. Create auto-chart for variable Target Gift Flag. How many donors are there in this file?
2. Add new canvas tab and estimate Decision Tree model with Target Gift Flag as Response. For Predictors use all variables, except of the following four: Control Number, Demographic Cluster, Target Gift Amount and Target Gift Amount with Zero. Rename tab from Page 2 to Decision Tree. Answer following questions:
   1. What is misclassification rate (event)?
   2. How many observations were used for Decision Tree model?
   3. Copy and paste decision tree. What variable is used in root node?
   4. Using Variable Importance plot, identify three most important variables in the decision tree?
   5. In the Options change Plot Type from the Variable Importance to Leaf Statistics. Which leaves have more than 80% customers who donated to veterans association?
   6. In Assessment Plots create Misclassification Plot. Does model predict better those who donated or not donated? (paste copy of that plot).
   7. In the Options menu change Prediction cut-off value -> increase to .8. Does it improve prediction of those who donated? (paste copy of that plot).
   8. In the Options menu change Prediction cut-off value -> decrease to .3. Does it improve prediction of those who donated? (paste copy of that plot).
3. Save report as PVA ML.