绪论 课程介绍和要求 课前测验

1、 问题:在其他因素相同的条件下,斜率估计量标准差较小,如果()
选项:
A: 解释变量有更多变差
B: 样本容量更小
C: 误差项的方差更大
D: 截矩估计值更小
答案: 【 解释变量有更多变差

2、 问题:在回归方程 = 698.9 – 2.28 × STR 中,如果斜率系数的 t- 统计量为 -4.38, 则它的标准误是()?
选项:
A:1.96
B:-1.96
C:0.52
D:4.38
答案: 【0.52

3、 问题:如果回归模型中遗漏了能够影响因变量的变量,会产生的后果是
选项:
A:一定会使得当前模型的最小二乘估计量有偏
B:既然其他变量没有包括进来,所以当前模型的估计是正确的
C:如果遗漏的变量和现存的变量相关,会使得当前的最小二乘估计量有偏
D:虽然无法度量出遗漏变量的作用,但是对模型中现存的变量进行估计不受影响
答案: 【如果遗漏的变量和现存的变量相关,会使得当前的最小二乘估计量有偏

4、 问题:为了判断哪个模型更好地拟合了数据,我们不能使用的原因是
选项:
A:当0会是负数
B:在这两个模型中,SST的单位不一样
C:在对数线性模型中,斜率系数的含义发生了变化
D:在对数线性模型中,可能会大于1
答案: 【在这两个模型中,SST的单位不一样

5、 问题:在一个包含两个变量的回归模型中,如果遗漏其中一个变量
选项:
A:如果遗漏变量和变量之间是负相关,不会影响前的系数估计值
B:一定会使的系数估计值上偏
C:即使在原来包含两个变量的回归中两个斜率系数都显著为正,也可能使变量前的系数估计值为负
D:将使变量和残差项的乘积的和不为0
答案: 【即使在原来包含两个变量的回归中两个斜率系数都显著为正,也可能使变量前的系数估计值为负

6、 问题:当经典线性回归模型去掉残差服从正态分布的假设时,仍然可以使用最小二乘法来估计未知参数,但是这时检验某个参数是否等于0的统计量不再服从t分布。
选项:
A:正确
B:错误
答案: 【错误

7、 问题:在一个研究工人年龄(Age)对时薪(AHE)的一元回归方程中,我们分别对具有高中文凭和大学文凭的工人做回归,得到如下估计结果:高中组: = 6.52 + 0.30 ´ Age (1.25) (0.04)大学组: = -4.44 + 0.92 ´ Age (1.77) (0.06)则大学组与高中组年龄对时薪的边际影响差异为()。(答案精确到小数点后两位)
答案: 【0.62

8、 问题:在上述两个回归中,假设两组样本是独立的,则大学组与高中组年龄对时薪的边际影响差异的置信区间上界为()。(答案精确到小数点后两位,。提示,对于两组独立样本,
答案: 【0.76

9、 问题:假设某研究者基于100个班的班级规模(CS)和平均测试成绩(TestScore)数据估计的OLS回归为:这100个班的测试成绩的样本标准差()等于( )(答案保留小数点后两位)
答案: 【11.90

10、 问题:具有两个自变量的同方差回归模型中,无约束的等于0.4366,如果加入两个约束条件再做回归得到有约束的分别等于0.4149. 已知样本观测个数为420,则F统计量为:(保留两位小数)
答案: 【8.03

第三章 经典线性回归模型 Quiz 2

1、 问题:The OLS estimator does NOT exist if
选项:
A: is singular
B:
C:for some
D:All of the above
答案: 【 is singular

2、 问题:If Assumption 3.1,3.2 and Assumption 3.3(a) and 3.4 holds, which of the following is NOT true about the OLS estimator ?
选项:
A:
B:
C:
D:
答案: 【

3、 问题:Which of the following is NOT true about the projection matrix ?
选项:
A:P is symmetric
B:
C: is also symmetric
D:
答案: 【

4、 问题:Which of the following is NOT true about Assumption 3.2 ?
选项:
A:Assumption 3.2 implies
B:Assumption 3.2 implies, for t = 1, 2, · · · , n.
C:Assumptions 3.1 and 3.2 ensure correct model specification on for a linear regression model.
D:Assumptions 3.1 and 3.2 can NOT guarantee correct model specification on for a linear regression model.
答案: 【Assumptions 3.1 and 3.2 can NOT guarantee correct model specification on for a linear regression model.

5、 问题:Which of the following is not true about ?
选项:
A: is always between 0 and 1
B:
C:When X contains an intercept term, is between 0 and 1
D:When X does not contain an intercept term, might be negative
答案: 【When X does not contain an intercept term, might be negative

6、 问题:When X is NOT stochastic, is equivalent to .
选项:
A:正确
B:错误
答案: 【正确

7、 问题:If is an independent random sample, is equivalent to for t = 1, 2, · · · , n.
选项:
A:正确
B:错误
答案: 【正确

8、 问题:Assumption 3.4 can be written more compactly as .
选项:
A:正确
B:错误
答案: 【错误

9、 问题:Let be the estimated residual from the OLS estimation. Wealways have .
选项:
A:正确
B:错误
答案: 【错误

10、 问题:Let be the estimated residual from the OLS estimation. ’s are independent.
选项:
A:正确
B:错误
答案: 【错误

11、 问题:Let be the estimated residual associated with the OLS estimator . We have under some assumptions.
选项:
A:正确
B:错误
答案: 【正确

12、 问题:In Section 3.7, under and some assumptions, we have .
选项:
A:正确
B:错误
答案: 【正确

13、 问题:In Section 3.7, under and some assumptions, we have .
选项:
A:正确
B:错误
答案: 【错误

14、 问题:When , where V is some known symmetric and positive definite matrix, in general.
选项:
A:正确
B:错误
答案: 【错误

15、 问题:When , where V is some known symmetric and positive definite matrix, we can still use the T and F test statistics derived in Section 3.7 to conduct hypothesis testing.
选项:
A:正确
B:错误
答案: 【错误

第五章 非独立样本的线性回归模型 Quiz 4

1、 问题:Which of the following is NOT true about strict stationarity?
选项:
A: If{Zt}is strictly stationary, the conditional probability of given past information set It−1 would have a time-invariant functional form.
B: No moment conditon on {} is needed to define strict stationarity.
C:When E() < ∞, Strict stationarity implies weak stationarity.
D: None of the above.
答案: 【 None of the above.

2、 问题:Which of the following assumption is NOT true about weak stationary time series process {}?
选项:
A:E() = µ for all t.
B:var() = < for all t.
C: Cov(,) depends on t.
D: All of the above.
答案: 【 Cov(,) depends on t.

3、 问题:Which of the following is true about ergodicity?
选项:
A:Ergodicity is a notion of asymptotic independence.
B:A strictly stationary process that is ergodic is called ergodic stationary.
C:An important implication of ergodicity is that the statistical properties (such as the population mean and variance) of the ergodic time series process can be deduced from a single, sufficiently long sample (realization) of the process.
D:All of the above.
答案: 【All of the above.

4、 问题:Which of the following is NOT true about a martingale difference sequence with {}?
选项:
A: can be interpreted as the difference of a martingale process.
B:E() = 0 for all t.
C:E( |) = 0.
D: None of the above.
答案: 【E() = 0 for all t.

5、 问题:Which of the following is NOT true about a white noise process {} ?
选项:
A:E() = 0
B:var() = < for all t.
C: Cov(,) depends on t.
D: All of the above.
答案: 【 Cov(,) depends on t.

6、 问题: Weak stationarity implies strict stationarity.
选项:
A:正确
B:错误
答案: 【错误

7、 问题: In Section 5.6, under and conditional homoskedasticity, we have (R -r)XN(0,R).
选项:

剩余60%内容付费后可查看
   

发表评论

电子邮件地址不会被公开。 必填项已用*标注