Leverage depends only on a point's x values, not its response. A point at an extreme x has high leverage, measured by its hat value; the average hat value is p/n for p model terms and n points, and values above about 2p/n deserve a look. High leverage combined with a large residual makes a point influential, meaning that removing it changes the coefficients a lot, which Cook's distance quantifies.
That one lot at 400 watts has all the leverage. Take it out and the power slope goes flat.
Also heard as
- hat value
Related terms
Regression
Fitting an equation that predicts a response from one or more input variables, usually by least squares.
Outlier
Observation that sits far from the pattern of the rest of the data, either a genuine extreme or an error.
Residual
Difference between an observed value and the value a fitted model predicts for it.
R-squared
Share of the variation in a response that a regression model accounts for, on a scale from 0 to 1.