Residuals
, the gap between the observed and fitted values. Distinguish it from the unobservable error , the gap from the true line.
When to use
OLS Estimation makes the residuals sum to zero and be orthogonal to every regressor. Positive means the point lies above the line (the model under-predicted), as with the roughly g residual in Assignment 1.
See Lecture 2 and Lecture 13. Worked in Assignment 1.