Concepts
response variable - correct answer ✔✔a variable of primary interest
explanatory variable - correct answer ✔✔a variable used to study the response variable
count variable - correct answer ✔✔a quantitative variable usually valid on non-negative integers
continuous variable - correct answer ✔✔a real-valued quantitative variable
nominal variable - correct answer ✔✔a categorical/qualitative variable having categories without a
meaningful or logical order
ordinal variable - correct answer ✔✔a categorical/qualitative variable having categories with a
meaningful or logical order
supervised learning - correct answer ✔✔Statistical Learning Element: has a response variable
unsupervised learning - correct answer ✔✔Statistical Learning Element: no response variable
regression response variable - correct answer ✔✔Statistical Learning Element: quantitative response
variable
classification - correct answer ✔✔Statistical Learning Element: categorical response variable
parametric - correct answer ✔✔Method Property: functional form of f specified
, non-parametric - correct answer ✔✔Method Property: functional form of f not specified
prediction - correct answer ✔✔Method Property: output of f^
inference - correct answer ✔✔Method Property: comprehension of f
flexibility - correct answer ✔✔Method Property: f^'s ability to follow the data
interpretability - correct answer ✔✔Method Property: f^'s ability to be understood
Training data - correct answer ✔✔Observations used to obtain f^
Testing data - correct answer ✔✔Observations not used to train/obtain f^
Danger of choosing a form for f that is not close to the truth - correct answer ✔✔Disadvantage to
parametric methods
Need for an abundance of observations - correct answer ✔✔Disadvantage to non-parametric methods
Flexibility and _______ are typically at odds - correct answer ✔✔Interpretability
As flexibility increases, training MSE ______, but the test MSE follows a ______________ - correct
answer ✔✔decreases; u-shaped pattern
Low flexibility leads to a method with _____ (low/high) variance and ______ (low/high) bias - correct
answer ✔✔low; high
Scatterplots - correct answer ✔✔plots values of two variables to investigate their relationship