
参及评分标准
卷别: A卷
一、名词解释题(每小题 2 分,共 10 分)
1、Time Series Data(时间系列数据):A time series is a set of observations on the values that a variable takes at different times. (3分)
2. Unbiasedness(无偏性):The estimator of, say, is said to be unbiased if its average or expected value, , is equal to the true value of,. (3分)
3. Goodness of Fit(拟合优度): A measure of how “well” the sample regression line fits the data. (3分)
4. Point Estimator(点估计量):The estimator is said to be point estimator if it provides a single (point) estimate of. (2分)
5. Type Ⅰ Error (Ⅰ类错误): Reject the Null Hypothesis when it is, in fact, true. (3分)
二、问答题(每小题 5分,共 20 分)
1. Assumptions underline the method of least squares.
(1) (1分)
(2) (1分)
(3) X is non-stochastic (1分)
(4) (1分)
(5) (1分)
2. Given the assumptions of CLRM, the OLS estimators are best, linear, unbiased estimators (BLUE). (5分)
3.
| Model | Equation | Sope(=) | Easticity(=) |
| Linear | |||
| Log-linear | |||
| Log-lin | |||
| Lin-log |
4. (2分)
se= (257.5874) (0.0399) =0.8775 (3分)
三、证明题(第一小题5分,第二小题5分,第三小题15分,共 25 分)
1. Prove: (5分)
2. Prove:
(5分)
3. Prove:
(2分)
Linearity of where, (1分)
Consider any other linear estimator say
That is also unbiased so
(2分)
* (2分)
To relate to , let
(2分)
and recall that
(2分)
Replace in * by and use
(4分)
has minimum variance.
四、综合应用题(第一小题 15 分,第二小题25分,共 40分)
1.
(1) (4分)
Where represents quantity of car sold. to captures the factors of quarters. is a dummy represent the regions in the country. is the random disturbance captures the factors that affect quantity of cars sold but out of the model.
=1 if the quantity of cars sold in quarter 1
=0 otherwise
=1 if the quantity of cars sold in quarter 2
=0 otherwise
=1 if the quantity of cars sold in quarter 3
=0 otherwise
=1 if the quantity of cars sold in the South of the country
=0 otherwise
(2) I use three dummy variables to represent the four quarters in a year and one dummy to represents the two regions in a country. We can only introduce m-1 dummy variables if a qualitative variable has m categories.
(4分)
(3) If I use 4 dummy variables to indicate 4 quarters in a year and also an intercept in the model, perfect collinearity would be result. However, if I use 4 dummy variables in absence of intercept, nothing will happen. (4分)
(4) Quantity of cars sold in north in quarter 4. (3分)
2.
(1)(5分)
t= (16.34) (-9.01) (-2.00) (-0.78) (0.96) (2.56)
P=(0.000) (0.000) (0.046) (0.438) (0.338) (0.011)
(2)(5分) keep other constant, as number of totwrk increased by 1, the sleep will decrease by .163 minutes; keep other constant, as a person receive 1 more years of schooling, he tends to sleep about 11 min less per week; keep other constant, as people get 1 year older, he tends to sleep about 9 min less per week; keep other constant, a male tends to sleep 88 min more than female per week.
(3) (5分)
Since the estimated is -.1634 is located in the 95% confidence interval (-.199,-.128), therefore reject and conclude that totwrk significantly affects child’s education at 95% confidence interval.
(4) (5分) Not all slope coefficients are simultaneously 0
Since P=0.000 for the test, thus reject and conclude that not all slope coefficients are simultaneously 0.
(5) (5分)means that about 12 percent of the variation in the sleep time per week is explained by totwrk, educ,age, agesq,male.
