Consider the 2013 declined loan data from LendingClub titled "RejectStatsB2013". Similar to the analysis done in the chapter, let's scrub the risk score data. First, since our analysis requires risk scores, debt-to-income data and employment length, we need to make sure each of them has valid data.

Required:

After removing the observations with a zero or missing risk score, which group (Excellent, Very Good, Good, Fair, Poor, Bad) had the most observations? Which had the least?