Lecture 4

Question 1

What happens if we have outliers when we are using the mean?

Answer 1

the mean is pulled in the direction of outliers

Question 2

What happens if we have outliers when we are using the median?

Answer 2

the median is robust to outliers

Question 3

What are the mean and median relative to the distribution?

Answer 3

they are “balance points”

Question 4

What minimizes the average L_infinity loss?

Answer 4

the midrange

Question 5

As p approaches infinity, the minimizer of mean L_p loss approaches what?

Answer 5

the midpoint of the minimum and maximum values in the dataset

Question 6

What does R_p(h) do as p approaches infinity?

Answer 6

minimizes the “worst case” distance from any data point

Question 7

When is the midrange the best prediction?

Answer 7

if your measure of “good” is “not far from any one data point”

Question 8

What is h* for 0-1 loss?

Answer 8

the mode

Question 9

What is the equation for the simple linear regression model?

Answer 9

H(x) = w_0 + w_1x