Question: 1. Since the variance of the regression coefficient β^ varies inversely with the variance of x, it is often suggested that we should drop all the observations in the middle range of x and use only the extreme observations on x in the calculation of β^. Is this a desirable procedure?
2. Suppose you are attempting to build a model that explains aggregate savings behavior as a function of the level of interest rates(IR). Would you rather sample during a period of fluctuating interest rates or a period of stable interest rates? Explain.