Applied Statistics - 3.1 - Outliers Flashcards
(4 cards)
1
Q
What are they, how to calculate
Outliers
A
- An extreme value that lies outside the overall pattern of the data
- Over Q₃: Q₃+ 1.5(Q₃ - Q₁)
- Under Q₁ : Q₁ - 1.5(Q₃ - Q₁)
2
Q
Why is it important to keep outliers in data?
A
- Outliers may indicate natural variation and is still a piece of data to keep
- May be the result of errors in measuring or recording data
3
Q
What is Cleaning the Data?
A
Removing anomalies from a data set
4
Q
How to calculate outliers using standard deviation
A
- Calculate mean
- Calculate standard deviation (square root of variance)
- Mean +/- standard deviation = outliers