Running the Model You will run the Predictive Choice model in Enginus. Within the model prompt: 1. Select “Choice between 2 alternatives (0/1)" under Target Variable. 2. Choose "Sundance Sample Data" from the dropdown box for Calibration Data. 3. Choose "Misbehaving Choice (No/Yes)" from the dropdown box for target variable. 4. Do not select "Box-Cox transform the predictors". 5. Choose "20-fold (most accurate)" for Cross-Validation. 6. Select the "Apply predictive model to out-of-sample data" button and choose "Sundance Out-of-Sample Data" from the dropdown box. Questions (2.5 points apiece) 1. On average, how much money did the in-sample customers from the dataset spend on purchases at Sundance? How many purchases did they make over the last three years? A. $95.30; 3.199 B. $200.92; 7.841 C. $203.00; 12.314 D. $474.00; 3.027 E. $200.92; 12.314 F. $95.30; 12.314 G. $203.00; 7.841 broc H. $474.00; 3.199 clood 2. Is the in-sample data skewed towards males or females? What about the out-of-sample data? How do we know for each? 3. How many customers in the calibration data purchased Misbehaving? Didn't purchase? Purchased: No Purchase:

Suppose, as an economist, you are asked to analyze an issue unlike anything you have ever done before. Also, suppose you do not have a specific model for analyzing that issue. What should you do? Hint: What would a carpenter do in a similar situation?

What is the difference between a positive and a normative statement?

Exercise B1 What point is preferred along an indifference Curve?