New Age Solar sells and installs solar panels for residential homes. The company’s sales representatives contact and pay a personal visit to potential customers to present the benefits of installing solar panels. This high-touch approach works well as the customers feel that they receive personal services that meet their individual needs, but it is more expensive than other, mass-marketing approaches. The company wants to be very strategic about visiting potential customers who are more likely to install solar panels. The company has compiled a data set of past home visits by sales reps. The data include the age (Age in years) and annual income (Income, in $1,000s) of the potential customer and whether or not the customer purchases the solar panels (Install: Y/N). A portion of the Solar_Data worksheet is shown in the accompanying table.

a. Create a bagging ensemble classification tree model to determine whether a customer purchases the solar panels. What are the overall accuracy rate, sensitivity, and specificity of the model on the validation data? 

b. Create a boosting ensemble classification tree model. What are the overall accuracy rate, sensitivity, and specificity of the model on the validation data? 

c. Compare the two ensemble models. Which model shows more robust performance according to the lift value of the leftmost bar of the decile-wise lift chart? 

d. Score the new cases in the Solar_Score worksheet using the bagging ensemble classification tree model. What percentage of the potential customers are likely to purchase the solar panels based on a cutoff probability value of 0.5?

Transcribed Image Text:

Income 115 68 1 73 Age 45 31 ⠀ 34 Install N Y # N