: You have created a model using mean imputation. At test time, you should fill in missing values with: View
: Which decision boundary corresponds to the following decision tree? In the options, red indicates high risk, blue indicates low risk. View
: Let’s say blood pressure (BP) measurements are more likely to be missing among young people, who generally have lower blood pressure. You use mean imputation to train your model. Which option correctly characterizes the mean BP (after imputation) in your training dataset? View
: True or False: When your data is missing at random, then whether or not you are missing a covariate is completely independent of your outcome. View
: You’ve fit a random forest of 10 trees with max depth 20. Your training ROC is 0.99 and test ROC is 0.54. Which of the following is NOT a reasonable thing to try? View
: One way to aggregate predictions from multiple trees is by a majority vote. Using this aggregation rule, select the prediction of the following trees on the data point (x=4, y=7, z=2): View
: Assume you have missing data on one of your features, and are considering two options: True or False: “Both options have the same performance”. View
: Using regression imputation, and the decision tree shown here, what is your prediction for this person’s risk of heart attack? View