Statistics Basics

Question 1

What are the three measures of centrality?

Answer 1

Mean, Median, Mode

Question 2

Define arithmetic mean

Answer 2

Average

Use for symmetric and unimodal data

Question 3

Define median

Answer 3

“the center value”
After sorting, the value at n/2 for data of size n
If n is even, take either of the middle values (or average of if possible)

Question 4

Define mode

Answer 4

Most common entry

Not very useful, mostly just an artefact of the data

Question 5

Define geometric mean

Answer 5

Average

Use for ratios or other values that are meant to be multiplied or are exponential in nature

Question 6

Define harmonic mean

Answer 6

Average
Use for average rate
Combine precision and recall for classifiers

Question 7

What are the two measures of variance?

Answer 7

Variance, standard deviation

Question 8

Define variance

Answer 8

Measurement of spread

high variance = data more spread out around average/mean

Question 9

Define standard deviation

Answer 9

sqrt(variance)