7.2.1博彩支出一例的方差和标准误差的计算 Table 7-1 Computations for the lotto example Estimator Formula Answer Equation number () 6.4854 (7.10) √G2=√64864 2.5468 (7.11 ∑X2)20740625) var(b1) w2 nx-10 (64864)93168 (7.12) selb. √ar(b1)=√93168 3.0523 (7.13 6.4864 var (b2) 0.000126 (7.14 ∑515602 se(b2) vwar(b2)=v0.00016 0.0112 (7.15) Note: The raw data underlying the calculations are given in Table 6. 4. In computing the variances of the estimators, o2 has been replaced by its estimator, 2. Note that the answers are rounded to the nearest digi 7-11
7-11 7.2.1 博彩支出一例的方差和标准误差的计算 Table 7-1 Computations for the lotto example
730LS估计量的性质 1.满足 Gauss-Markoy theorem 如果满足经典线性回归模型的基本假设,则在所有 的线性估计量中,OLS估计量是BLUE。 2.0LS估计量的性质: 参数的估计量是线性的;无偏的;有效的 随机误差项的方差的估计量是无偏的; 7-12
7-12 7.3 OLS估计量的性质 1. 满足Gauss-Markov theorem 如果满足经典线性回归模型的基本假设，则在所有 的线性估计量中， OLS估计量是BLUE。 2. OLS估计量的性质： 参数的估计量是线性的；无偏的；有效的； 随机误差项的方差的估计量是无偏的；
Monte carlo试验 Table 7-2 Monte Carlo experiment: Y, 1.5+ 2X+;uNo, 4) b 2247 1.840 27159 0360 2.090 7.1663 2483 2.558 3.3306 0220 2.180 2.0794 3.070 1620 4.3932 2.570 1830 7.1770 2.551 1.928 5.7552 0.060 2070 3.6176 2.170 2.537 3.4708 1470 2.020 4.4479 2.540 1.970 2.1756 2340 1.960 2.8291 0.775 2.050 15252 3.020 1.740 15104 0810 1.940 4.7830 1890 1.890 7.3658 2760 1.820 18036 2.130 18796 0950 2.030 4.9908 2960 1.840 4.5514 3430 1.740 52258 7-13 b1=14526b2=196652=44743
7-13 Monte Carlo试验 Table 7-2 Monte Carlo experiment: Yi =1.5 + 2Xi +ui; u ~ N(0, 4)