上机课实验作业三姓名：李姝仪学号：2012310320班级：12级审计二班一：数据集DATA4-71）先验地预期CM和各个变量之间的关系，并计算样本相关系数。答：先验地预期：CM与FLR之间呈负相关关系，即女性文盲率越高，婴儿死亡率越低；CM与PGNP之间呈负相关关系，即人均国民产值越高，婴儿死亡率越低；CM与TFR之间呈正相关关系，即总生育率越高，婴儿死亡率就越高。对应的样本相关系数如下表所示：.correlatecmflrpgnptfr(obs=64)|cmflrpgnptfr-------------+------------------------------------cm|1.0000flr|-0.81831.0000pgnp|-0.40770.26851.0000tfr|0.6711-0.6260-0.18571.00002）做CM对FLR的回归。答：输入stata命令：.regcmflrSource|SSdfMSNumberofobs=64-------------+------------------------------F(1,62)=125.65Model|243515.0491243515.049Prob>F=0.0000Residual|120162.951621938.11211R-squared=0.6696-------------+------------------------------AdjR-squared=0.6643Total|363678635772.66667RootMSE=44.024------------------------------------------------------------------------------cm|Coef.Std.Err.tP>|t|[95%Conf.Interval]-------------+----------------------------------------------------------------flr|-2.390496.2132625-11.210.000-2.816802-1.96419_cons|263.863512.2249921.580.000239.4261288.3009得回归结果：CM=263.86-2.39FLR。3）做CM对FLR和PGNP的回归。答：输入stata命令：.regcmflrpgnpSource|SSdfMSNumberofobs=64-------------+------------------------------F(2,61)=73.83Model|257362.3732128681.187Prob>F=0.0000Residual|106315.627611742.87913R-squared=0.7077-------------+------------------------------AdjR-squared=0.6981Total|363678635772.66667RootMSE=41.748------------------------------------------------------------------------------cm|Coef.Std.Err.tP>|t|[95%Conf.Interval]-------------+----------------------------------------------------------------flr|-2.231586.2099472-10.630.000-2.651401-1.81177pgnp|-.0056466.0020033-2.820.006-.0096524-.0016408_cons|263.641611.5931822.740.000240.4596286.8236得回归结果：CM=263.64-2.23FLR-0.0056PGNP。4）做CM对FLR、PGNP和TFR的回归。观察校正拟合优度的变化。答：输入stata命令：.regcmflrpgnptfrSource|SSdfMSNumberofobs=64-------------+------------------------------F(3,60)=59.17Model|271802.616390600.8721Prob>F=0.0000Residual|91875.3836601531.25639R-squared=0.7474-------------+------------------------------AdjR-squared=0.7347Total|363678635772.66667RootMSE=39.131------------------------------------------------------------------------------cm|Coef.Std.Err.tP>|t|[95%Conf.Interval]-------------+----------------------------------------------------------------flr|-1.768029.2480169-7.130.000-2.264137-1.271921pgnp|-.0055112.0018782-2.930.005-.0092682-.0017542tfr|12.868644.1905333.070.0034.48632321.25095_cons|168.306732.891665.120.000102.5136234.0998得回归结果：CM=168.31-1.77FLR-0.0055PGNP+12.87TFR。观察发现校正拟合优度随着解释变量个数的增加而不断增大，但始终小于拟合优度的数值。5）根据各种回归结果，选择哪个模型？为什么？答：根据以上回归结果，选择4）中的模型。因为此模型中解释变量个数最多，考虑的变量因素多，且此模型的拟合优度和校正拟合优度都比前几个模型大，说明此模型对因变量的解释力较前几个模型更好些，得到的结果更准确。6）对3）中的回归，检验FLR和PGNP的联合显著性。（写出原假设、备择假设、检验统计量）答：输入stata命令：.regcmflrpgnp（结果略）.testflrpgnp(1)flr=0(2)pgnp=0F(2,61)=73.83Prob>F=0.0000其中：原假设：H0：β2=β3=0备择假设：H1：β2与β3至少有一个不为零。检验统计量：F值，F(2，61)=73.83，且Prob>F=0.0000，说明FLR和PGNP通...
