Chapter 8

Extrapolation

Making predictions outside the range of observed data, which can be unreliable as it assumes the established pattern continues beyond the known data.

Outlier

An unusual data point that differs significantly from the other data points, potentially affecting the results of a regression analysis.

Leverage

The degree to which a data point can influence the position of the regression line, with points far from the mean of the predictor variable having higher leverage.

Influential Point

A data point that significantly changes the result of the regression analysis when it is included or excluded, due to its position and leverage.

Lurking Variable

A variable not included in the analysis but which influences both the explanatory and response variables, potentially confounding the results.

Re-Expression

Transforming data using mathematical functions to make the relationship between variables more linear and satisfy the assumptions of regression.

Ladder Of Powers

A sequence of transformations applied to data, ranging from roots and logarithms to powers, to find the best re-expression for linearizing relationships and stabilizing variance.