c) For simple linear regression, the standard error of the slope estimator is √MSE/Sx, where sx is the standard deviatio
Posted: Fri Jul 01, 2022 5:28 am
c) For simple linear regression, the standard error of the slope estimator is √MSE/Sx, where sx is the standard deviation of the X-values. So, the estimate is most accurate if X-values are maximally spread. If X € [a, b], maximal sx occurs when half the X-values are equal to a and the other half equal to b. Is it a good idea to sample this way? [HINT: Consider the residual plot.] [4 marks]