then your friends data have the variance, varO/i) = 02(1 + 3c), for some constant c 2 0. (d) Your friend is sorry that % of the data are not as precise as they could have been, and suggest that you discard the noise data, and simply use r12 = 21%?) 21231\" Y; as your estimator for [1.17. Which estimator is most efcient (has the smallest variance) {11 or g? Does your answer depend on c? (e) Suppose now that c = 0 such that var(Y,-) = o2 for 3' = 1, . . .,N. You A 2 have N = 100 observation and calculate 82 = 2:1 (Y; [11) 2 50,000,000 and [111 = $45, 300. Before collecting the data, your friend argues mean salary, y, is $46, ZZZ, using a 5% signicance level. In