Return to the flight delays problem above and run it again, only this time choose a boosted classification tree. Leave the default number of weak learners, and select re-sampling. Set maximum levels to display at 6 , and minimum number of records in a terminal node to 1.

a. Compared with the single tree, how does the boosted tree behave in terms of overall error rate?

b. Compared with the single tree, how does the boosted tree behave in terms of error rate in identifying delayed flights?

c. Explain why this model might have the best performance over the other models you fit.