Sometimes it is believed that 0 is 0 because we think that E(Y | X = 0) = 0. Then the appropriate model is yi

Sometimes it is believed that β0 is 0 because we think that E(Y | X = 0) = 0. Then the appropriate model is
yi = β1Xi + ˆˆi:
This model is usually called "regression through the origin" since the regression line is forced through the origin. The least-squares estimator of β1 minimizes

(y- ,}2, i=1