Residuals
Residual is the y-axis distance between where a point actually is and where the linear regression line crosses the y-axis for that x-value.
- E.g. if the point was (3,8) and the linear regression line crosses at the point (3,10) then the Residual == -2
- Residuals help us measure the line of best fit through a scatterplot
- \(r_i = y_i - \hat{y}\)
- Reads: The residual of a point i equals the actual y value at point i, minus the predicted regression line value at point i
Graph:
- 113.033 Statistics - Residuals to 113.031 Statistics - Linear Regression
- 113.033 Statistics - Residuals to 113.032 Statistics - Linear Regression Least Squares method
- 113.033 Statistics - Residuals to 113.034 Statistics - Coefficient of determination and Squared error
- 113.033 Statistics - Residuals to 113.035 Statistics - Standard deviation of residuals
- 113.032 Statistics - Linear Regression Least Squares method to 113.033 Statistics - Residuals
- 113.035 Statistics - Standard deviation of residuals to 113.033 Statistics - Residuals