A fellow researcher fits a simple linear regression model Yi = α + βXi + ui by least squares, but hasdoubts that the first least squares assumption E(ui |Xi) = 0 is correct for her data set. She decides totest for this assumption by calculating the residuals from her regression ûi = Yi – Ŷi and thenperforming the usual t-test of the null hypothesis that their (population) mean is zero, assuming theerrors are normally distributed. Evaluate her strategy.

A fellow researcher fits a simple linear regression model Yi = α + βXi + ui by least squares, buthasdoubts that the first least squares assumption E(ui |Xi) = 0 is correct for her data set….