1. Outliers, leverage and influential points
Warm-up
Three quick questions from earlier lessons.
- , , , , .
- The slope :
- The intercept :
- A point is in the data. Its residual:
- The fraction of the variation in explained by the line ():
- . A data point is . Find its residual.
- The residual plot of a linear model fitted to 56 points shows a clear U-shaped curve. What does that tell you?
- A straight line is not a good model: the relationship is curved
- The model is good, because residuals are both positive and negative
- There is one outlier
Discover
A scatterplot of 20 points has a line through them. One more point is added far to the right, well below the trend. What happens to the line?
- It tilts towards the new point
- Nothing: one point cannot change a line fitted to 20
Key idea
Some unusual points matter more than others.
Worked example
Three unusual points.
Your turn, guided
Your turn: removing a point changes the slope from 2.1 to 0.4. By how much did the slope change?
Practice ★
- In a scatterplot, the -values lie between 10 and 51, except one point at . What is that point called?
- A high-leverage point: its x is far from the others
- A point with a large residual
- The median point
- Removing one of 52 points changes the regression slope a lot. What is that point?
- An influential point
- Any outlier: all outliers change the slope a lot
- A typical point near the center of the data
Watch out!
Practice ★★
- Among 46 points, one has close to but a very large residual. Removing it mostly changes what?
- The correlation (it rises); the slope barely changes
- The slope a lot
- Nothing at all
- Among 77 points, one has close to but a very large residual. Removing it mostly changes what?
- The correlation (it rises); the slope barely changes
- The slope a lot
- Nothing at all
Exit ticket
Three questions to check you have got it.
- In a scatterplot, the -values lie between 10 and 51, except one point at . What is that point called?
- A high-leverage point: its x is far from the others
- A point with a large residual
- The median point
- In a scatterplot, the -values lie between 10 and 36, except one point at . What is that point called?
- A high-leverage point: its x is far from the others
- A point with a large residual
- The median point
- In a scatterplot, the -values lie between 10 and 30, except one point at . What is that point called?
- A high-leverage point: its x is far from the others
- A point with a large residual
- The median point
Summary card
Unusual points
- Outlier: large residual. Leverage: extreme .
- Influential: removing it changes the line.
Watch out: leverage, not residual size, drives influence.
One more step is optional: a challenge. Try it, or finish the lesson now.
Challenge ★★★
- Removing one influential point changes the correlation from to . By how much does increase?