第二章 简单线性回归模型
2.1
(1) ①首先分析人均寿命与人均 GDP的数量关系,用 Eviews 分析:
Dependent Variable: Y
Method: Least Squares
Date: 12/27/14 Time: 21:00
Sample: 1 22
Included observations: 22
Variable Coefficient Std. Error t-Statistic Prob.
C 56.64794 1.960820 28.88992 0.0000
X1 0.128360 0.027242 4.711834 0.0001
R-squared 0.526082 Mean dependent var 62.50000
Adjusted R-squared 0.502386 S.D. dependent var 10.08889
S.E. of regression 7.116881 Akaike info criterion 6.849324
Sum squared resid 1013.000 Schwarz criterion 6.948510
Log likelihood -73.34257 Hannan-Quinn criter. 6.872689
F-statistic 22.20138 Durbin-Watson stat 0.629074
Prob(F-statistic) 0.000134
有上可知,关系式为 y=56.64794+0.128360x 1
②关于人均寿命与成人识字率的关系,用 Eviews 分析如下:
Dependent Variable: Y
Method: Least Squares
Date: 11/26/14 Time: 21:10
Sample: 1 22
Included observations: 22
Variable Coefficient Std. Error t-Statistic Prob.
C 38.79424 3.532079 10.98340 0.0000
X2 0.331971 0.046656 7.115308 0.0000
R-squared 0.716825 Mean dependent var 62.50000
Adjusted R-squared 0.702666 S.D. dependent var 10.08889
S.E. of regression 5.501306 Akaike info criterion 6.334356
Sum squared resid 605.2873 Schwarz criterion 6.433542
Log likelihood -67.67792 Hannan-Quinn criter. 6.357721
F-statistic 50.62761 Durbin-Watson stat 1.846406
Prob(F-statistic) 0.000001
由上可知,关系式为 y=38.79424+0.331971x 2
③关于人均寿命与一岁儿童疫苗接种率的关系,用 Eviews 分析如下:
Dependent Variable: Y
Method: Least Squares
Date: 11/26/14 Time: 21:14
Sample: 1 22
Included observations: 22
Variable Coefficient Std. Error t-Statistic Prob.
C 31.79956 6.536434 4.864971 0.0001
X3 0.387276 0.080260 4.825285 0.0001
R-squared 0.537929 Mean dependent var 62.50000
Adjusted R-squared 0.514825 S.D. dependent var 10.08889
S.E. of regression 7.027364 Akaike info criterion 6.824009
Sum squared resid 987.6770 Schwarz criterion 6.923194
Log likelihood -73.06409 Hannan-Quinn criter. 6.847374
F-statistic 23.28338 Durbin-Watson stat 0.952555
Prob(F-statistic) 0.000103
由上可知,关系式为 y=31.79956+0.387276x 3
(2)①关于人均寿命与人均 GDP 模型,由上可知,可决系数为 0.526082 ,说明所建模型
整体上对样本数据拟合较好。
对于回归系数的 t 检验: t(β 1)=4.711834>t 0.025 (20)=2.086 ,对斜率系数的显著性检验
表明,人均 GDP 对人均寿命有显著影响。
②关于人均寿命与成人识字率模型,由上可知,可决系数为 0.716825 ,说明所建模型整体
上对样本数据拟合较好。
对于回归系数的 t 检验: t(β 2)=7.115308>t 0.025 (20)=2.086 ,对斜率系数的显著性检验表
明,成人识字率对人均寿命有显著影响。
③关于人均寿命与一岁儿童疫苗的模型,由上可知,可决系数为 0.537929 ,说明所建模型
整体上对样本数据拟合较好。
对于回归系数的 t 检验: t(β 3)=4.825285>t 0.025 (20)=2.086 ,对斜率系数的显著性检验
表明,一岁儿童疫苗接种率对人均寿命有显著影响。
2.2
(1)
①对于浙江省预算收入与全省生产总值的模型,用 Eviews 分析结果如下:
Dependent Variable: Y
Method: Least Squares
Date: 12/03/14 Time: 17:00
Sample (adjusted): 1 33
Included observations: 33 after adjustments
Variable Coefficient Std. Error t-Statistic Prob.
X 0.176124 0.004072 43.25639 0.0000
C -154.3063 39.08196 -3.948274 0.0004
R-squared 0.983702 Mean dependent var 902.5148
Adjusted R-squared 0.983177 S.D. dependent var 1351.009
S.E. of regression 175.2325 Akaike info criterion 13.22880
Sum squared resid 951899.7 Schwarz criterion 13.31949
Log likelihood -216.2751 Hannan-Quinn criter. 13.25931
F-statistic 1871.115 Durbin-Watson stat 0.100021
Prob(F-statistic) 0.000000
②由上可知,模型的参数:斜率系数 0.176124,截距为 —154.3063
③关于浙江省财政预算收入与全省生产总值的模型,检验模型的显著性:
1)可决系数为 0.983702 ,说明所建模型整体上对样本数据拟合较好。
2)对于回归系数的 t 检验: t(β2)=43.25639>t 0.025 (31)=2.0395 ,对斜率系数的显著性检
验表明,全省生产总值对财政预算总收入有显著影响。
④用
规范
编程规范下载gsp规范下载钢格栅规范下载警徽规范下载建设厅规范下载
形式写出检验结果如下:
Y=0.176124X —154.3063
(0.004072) (39.08196)
t= (43.25639) ( -3.948274 )
R2=0.983702 F=1871.115 n=33
⑤经济意义是:全省生产总值每增加 1 亿元,财政预算总收入增加 0.176124 亿元。
(2)当 x=32000 时,
①进行点预测,由上可知 Y=0.176124X —154.3063 ,代入可得:
Y= Y=0.176124*32000 —154.3063=5481.6617
②进行区间预测:
先由 Eviews 分析:
X Y
Mean 6000.441 902.5148
Median 2689.280 209.3900
Maximum 27722.31 4895.410
Minimum 123.7200 25.87000
Std. Dev. 7608.021 1351.009
Skewness 1.432519 1.663108
Kurtosis 4.010515 4.590432
Jarque-Bera 12.69068 18.69063
Probability 0.001755 0.000087
Sum 198014.5 29782.99
Sum Sq. Dev. 1.85E+09 58407195
Observations 33 33
由上表可知,
∑x
2
=∑(Xi—X)
2
=δ
2
x(n—1)= 7608.021 2 x (33—1)=1852223.473
(Xf—X)2=(32000— 6000.441)2=675977068.2
当 Xf=32000 时,将相关数据代入计算得到:
5481.6617— 2.0395x175.2325x √ 1/33+1852223.473/675977068.2 ≤
Yf ≤ 5481.6617+2.0395x175.2325x √ 1/33+1852223.473/675977068.2
即 Yf 的置信区间为( 5481.6617—64.9649, 5481.6617+64.9649 )
(3) 对于浙江省预算收入对数与全省生产总值对数的模型,由 Eviews 分析结果如下:
Dependent Variable: LNY
Method: Least Squares
Date: 12/03/14 Time: 18:00
Sample (adjusted): 1 33
Included observations: 33 after adjustments
Variable Coefficient Std. Error t-Statistic Prob.
LNX 0.980275 0.034296 28.58268 0.0000
C -1.918289 0.268213 -7.152121 0.0000
R-squared 0.963442 Mean dependent var 5.573120
Adjusted R-squared 0.962263 S.D. dependent var 1.684189
S.E. of regression 0.327172 Akaike info criterion 0.662028
Sum squared resid 3.318281 Schwarz criterion 0.752726
Log likelihood -8.923468 Hannan-Quinn criter. 0.692545
F-statistic 816.9699 Durbin-Watson stat 0.096208
Prob(F-statistic) 0.000000
①模型方程为: lnY=0.980275lnX-1.918289
②由上可知,模型的参数:斜率系数为 0.980275 ,截距为 -1.918289
③关于浙江省财政预算收入与全省生产总值的模型,检验其显著性:
1)可决系数为 0.963442 ,说明所建模型整体上对样本数据拟合较好。
2)对于回归系数的 t 检验: t(β 2) =28.58268>t 0.025 (31)=2.0395 ,对斜率系数的显著性检
验表明,全省生产总值对财政预算总收入有显著影响。
④经济意义:全省生产总值每增长 1%,财政预算总收入增长 0.980275%
2.4
(1)对建筑面积与建造单位成本模型,用 Eviews 分析结果如下:
Dependent Variable: Y
Method: Least Squares
Date: 12/01/14 Time: 12:40
Sample: 1 12
Included observations: 12
Variable Coefficient Std. Error t-Statistic Prob.
X -64.18400 4.809828 -13.34434 0.0000
C 1845.475 19.26446 95.79688 0.0000
R-squared 0.946829 Mean dependent var 1619.333
Adjusted R-squared 0.941512 S.D. dependent var 131.2252
S.E. of regression 31.73600 Akaike info criterion 9.903792
Sum squared resid 10071.74 Schwarz criterion 9.984610
Log likelihood -57.42275 Hannan-Quinn criter. 9.873871
F-statistic 178.0715 Durbin-Watson stat 1.172407
Prob(F-statistic) 0.000000
由上可得:建筑面积与建造成本的回归方程为:
Y=1845.475--64.18400X
(2)经济意义:建筑面积每增加 1 万平方米,建筑单位成本每平方米减少 64.18400 元。
(3)
①首先进行点预测,由 Y=1845.475--64.18400X 得,当 x=4.5 ,y=1556.647
②再进行区间估计:
用 Eviews 分析:
Y X
Mean 1619.333 3.523333
Median 1630.000 3.715000
Maximum 1860.000 6.230000
Minimum 1419.000 0.600000
Std. Dev. 131.2252 1.989419
Skewness 0.003403 -0.060130
Kurtosis 2.346511 1.664917
Jarque-Bera 0.213547 0.898454
Probability 0.898729 0.638121
Sum 19432.00 42.28000
Sum Sq. Dev. 189420.7 43.53567
Observations 12 12
由上表可知,
∑x
2
=∑(Xi—X)
2
=δ
2
x(n—1)= 1.989419 2 x (12—1)=43.5357
(Xf—X)2=(4.5— 3.523333 )2=0.95387843
当 Xf=4.5 时,将相关数据代入计算得到:
1556.647 —2.228x31.73600 x√1/12+43.5357/0.95387843 ≤
Yf ≤1556.647 +2.228x31.73600 x√ 1/12+43.5357/0.95387843
即 Yf 的置信区间为( 1556.647 —478.1231, 1556.647 +478.1231)
3.1
(1)
①对百户拥有家用汽车量计量经济模型,用 Eviews 分析结果如下:
Dependent Variable: Y
Method: Least Squares
Date: 11/25/14 Time: 12:38
Sample: 1 31
Included observations: 31
Variable Coefficient Std. Error t-Statistic Prob.
X2 5.996865 1.406058 4.265020 0.0002
X3 -0.524027 0.179280 -2.922950 0.0069
X4 -2.265680 0.518837 -4.366842 0.0002
C 246.8540 51.97500 4.749476 0.0001
R-squared 0.666062 Mean dependent var 16.77355
Adjusted R-squared 0.628957 S.D. dependent var 8.252535
S.E. of regression 5.026889 Akaike info criterion 6.187394
Sum squared resid 682.2795 Schwarz criterion 6.372424
Log likelihood -91.90460 Hannan-Quinn criter. 6.247709
F-statistic 17.95108 Durbin-Watson stat 1.147253
Prob(F-statistic) 0.000001
②得到模型得:
Y=246.8540+5.996865X 2- 0.524027 X 3-2.265680 X 4
③对模型进行检验:
1) 可决系数是 0.666062 ,修正的可决系数为 0.628957 ,说明模型对样本拟合较好
2) F 检验, F=17.95108>F (3,27 )=3.65 ,回归方程显著。
3)t 检验, t 统计量分别为 4.749476 ,4.265020 ,-2.922950 , -4.366842 ,均大于
t(27)=2.0518,所以这些系数都是显著的。
④依据:
1) 可决系数越大,说明拟合程度越好
2) F 的值与临界值比较,若大于临界值,则否定原假设,回归方程是显著的;若小于临界
值,则接受原假设,回归方程不显著。
3) t 的值与临界值比较,若大于临界值,则否定原假设,系数都是显著的;若小于临界值,
则接受原假设,系数不显著。
(2)经济意义:人均GDP增加1万元,百户拥有家用汽车增加 5.996865 辆,城镇人口
比重增加1个百分点,百户拥有家用汽车减少 0.524027 辆,交通工具消费价格指数每上升
1,百户拥有家用汽车减少 2.265680 辆。
(3)用 EViews 分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/08/14 Time: 17:28
Sample: 1 31
Included observations: 31
Variable Coefficient Std. Error t-Statistic Prob.
X2 5.135670 1.010270 5.083465 0.0000
LNX3 -22.81005 6.771820 -3.368378 0.0023
LNX4 -230.8481 49.46791 -4.666624 0.0001
C 1148.758 228.2917 5.031974 0.0000
R-squared 0.691952 Mean dependent var 16.77355
Adjusted R-squared 0.657725 S.D. dependent var 8.252535
S.E. of regression 4.828088 Akaike info criterion 6.106692
Sum squared resid 629.3818 Schwarz criterion 6.291723
Log likelihood -90.65373 Hannan-Quinn criter. 6.167008
F-statistic 20.21624 Durbin-Watson stat 1.150090
Prob(F-statistic) 0.000000
模型方程为:
Y=5.135670 X 2-22.81005 LNX 3-230.8481 LNX 4+1148.758
此分析得出的可决系数为 0.691952>0.666062 ,拟合程度得到了提高,可这样改进。
3.2
(1)对出口货物总额计量经济模型,用 Eviews 分析结果如下: :
Dependent Variable: Y
Method: Least Squares
Date: 12/01/14 Time: 20:25
Sample: 1994 2011
Included observations: 18
Variable Coefficient Std. Error t-Statistic Prob.
X2 0.135474 0.012799 10.58454 0.0000
X3 18.85348 9.776181 1.928512 0.0729
C -18231.58 8638.216 -2.110573 0.0520
R-squared 0.985838 Mean dependent var 6619.191
Adjusted R-squared 0.983950 S.D. dependent var 5767.152
S.E. of regression 730.6306 Akaike info criterion 16.17670
Sum squared resid 8007316. Schwarz criterion 16.32510
Log likelihood -142.5903 Hannan-Quinn criter. 16.19717
F-statistic 522.0976 Durbin-Watson stat 1.173432
Prob(F-statistic) 0.000000
①由上可知,模型为:
Y = 0.135474X 2 + 18.85348X 3 - 18231.58
②对模型进行检验:
1)可决系数是 0.985838 ,修正的可决系数为 0.983950 ,说明模型对样本拟合较好
2)F 检验, F=522.0976>F (2,15 )=4.77 ,回归方程显著
3)t 检验, t 统计量分别为 X2 的系数对应 t 值为 10.58454 ,大于 t( 15)=2.131,系数是显
著的, X3 的系数对应 t 值为 1.928512 ,小于 t(15)=2.131,说明此系数是不显著的。
(2)对于对数模型,用 Eviews 分析结果如下:
Dependent Variable: LNY
Method: Least Squares
Date: 12/01/14 Time: 20:25
Sample: 1994 2011
Included observations: 18
Variable Coefficient Std. Error t-Statistic Prob.
LNX2 1.564221 0.088988 17.57789 0.0000
LNX3 1.760695 0.682115 2.581229 0.0209
C -20.52048 5.432487 -3.777363 0.0018
R-squared 0.986295 Mean dependent var 8.400112
Adjusted R-squared 0.984467 S.D. dependent var 0.941530
S.E. of regression 0.117343 Akaike info criterion -1.296424
Sum squared resid 0.206540 Schwarz criterion -1.148029
Log likelihood 14.66782 Hannan-Quinn criter. -1.275962
F-statistic 539.7364 Durbin-Watson stat 0.686656
Prob(F-statistic) 0.000000
①由上可知,模型为:
LNY=-20.52048+1.564221 LNX 2+1.760695 LNX 3
②对模型进行检验:
1)可决系数是 0.986295 ,修正的可决系数为 0.984467 ,说明模型对样本拟合较好。
2)F 检验, F=539.7364> F (2,15 )=4.77 ,回归方程显著。
3)t 检验, t 统计量分别为 -3.777363 ,17.57789 ,2.581229 ,均大于 t(15)=2.131,所以
这些系数都是显著的。
(3)
①( 1)式中的经济意义:工业增加 1 亿元,出口货物总额增加 0.135474 亿元,人民币汇
率增加 1,出口货物总额增加 18.85348 亿元。
②( 2)式中的经济意义:工业增加额每增加 1%,出口货物总额增加 1.564221% ,人民币
汇率每增加 1% ,出口货物总额增加 1.760695%
3.3
(1)对家庭书刊消费对家庭月平均收入和户主受教育年数计量模型,由 Eviews 分析结果如
下:
Dependent Variable: Y
Method: Least Squares
Date: 12/01/14 Time: 20:30
Sample: 1 18
Included observations: 18
Variable Coefficient Std. Error t-Statistic Prob.
X 0.086450 0.029363 2.944186 0.0101
T 52.37031 5.202167 10.06702 0.0000
C -50.01638 49.46026 -1.011244 0.3279
R-squared 0.951235 Mean dependent var 755.1222
Adjusted R-squared 0.944732 S.D. dependent var 258.7206
S.E. of regression 60.82273 Akaike info criterion 11.20482
Sum squared resid 55491.07 Schwarz criterion 11.35321
Log likelihood -97.84334 Hannan-Quinn criter. 11.22528
F-statistic 146.2974 Durbin-Watson stat 2.605783
Prob(F-statistic) 0.000000
①模型为: Y = 0.086450X + 52.37031T-50.01638
②对模型进行检验:
1)可决系数是 0.951235 ,修正的可决系数为 0.944732 ,说明模型对样本拟合较好。
2)F 检验, F=539.7364> F (2,15 )=4.77 ,回归方程显著。
3)t 检验, t 统计量分别为 2.944186 ,10.06702 ,均大于 t(15)=2.131,所以这些系数都
是显著的。
③经济意义:家庭月平均收入增加 1 元,家庭书刊年消费支出增加 0.086450 元,户主受教
育年数增加 1 年,家庭书刊年消费支出增加 52.37031 元。
(2)用 Eviews 分析:
①
Dependent Variable: Y
Method: Least Squares
Date: 12/01/14 Time: 22:30
Sample: 1 18
Included observations: 18
Variable Coefficient Std. Error t-Statistic Prob.
T 63.01676 4.548581 13.85416 0.0000
C -11.58171 58.02290 -0.199606 0.8443
R-squared 0.923054 Mean dependent var 755.1222
Adjusted R-squared 0.918245 S.D. dependent var 258.7206
S.E. of regression 73.97565 Akaike info criterion 11.54979
Sum squared resid 87558.36 Schwarz criterion 11.64872
Log likelihood -101.9481 Hannan-Quinn criter. 11.56343
F-statistic 191.9377 Durbin-Watson stat 2.134043
Prob(F-statistic) 0.000000
②
Dependent Variable: X
Method: Least Squares
Date: 12/01/14 Time: 22:34
Sample: 1 18
Included observations: 18
Variable Coefficient Std. Error t-Statistic Prob.
T 123.1516 31.84150 3.867644 0.0014
C 444.5888 406.1786 1.094565 0.2899
R-squared 0.483182 Mean dependent var 1942.933
Adjusted R-squared 0.450881 S.D. dependent var 698.8325
S.E. of regression 517.8529 Akaike info criterion 15.44170
Sum squared resid 4290746. Schwarz criterion 15.54063
Log likelihood -136.9753 Hannan-Quinn criter. 15.45534
F-statistic 14.95867 Durbin-Watson stat 1.052251
Prob(F-statistic) 0.001364
以上分别是 y 与 T,X 与 T 的一元回归
模型分别是:
Y = 63.01676T - 11.58171
X = 123.1516T + 444.5888
(3)对残差进行模型分析,用 Eviews分析结果如下:
Dependent Variable: E1
Method: Least Squares
Date: 12/03/14 Time: 20:39
Sample: 1 18
Included observations: 18
Variable Coefficient Std. Error t-Statistic Prob.
E2 0.086450 0.028431 3.040742 0.0078
C 3.96E-14 13.88083 2.85E-15 1.0000
R-squared 0.366239 Mean dependent var 2.30E-14
Adjusted R-squared 0.326629 S.D. dependent var 71.76693
S.E. of regression 58.89136 Akaike info criterion 11.09370
Sum squared resid 55491.07 Schwarz criterion 11.19264
Log likelihood -97.84334 Hannan-Quinn criter. 11.10735
F-statistic 9.246111 Durbin-Watson stat 2.605783
Prob(F-statistic) 0.007788
模型为:
E1 = 0.086450E 2 + 3.96e-14
参数:斜率系数 α 为 0.086450,截距为 3.96e-14
(3)由上可知, β2 与α2 的系数是一样的。回归系数与被解释变量的残差系数是一样的,
它们的变化规律是一致的。
3.6
(1)预期的符号是 X1,X2,X3,X4,X5 的符号为正, X6 的符号为负
(2)根据 Eviews 分析得到数据如下:
Dependent Variable: Y
Method: Least Squares
Date: 12/04/14 Time: 13:24
Sample: 1994 2011
Included observations: 18
Variable Coefficient Std. Error t-Statistic Prob.
X2 0.001382 0.001102 1.254330 0.2336
X3 0.001942 0.003960 0.490501 0.6326
X4 -3.579090 3.559949 -1.005377 0.3346
X5 0.004791 0.005034 0.951671 0.3600
X6 0.045542 0.095552 0.476621 0.6422
C -13.77732 15.73366 -0.875659 0.3984
R-squared 0.994869 Mean dependent var 12.76667
Adjusted R-squared 0.992731 S.D. dependent var 9.746631
S.E. of regression 0.830963 Akaike info criterion 2.728738
Sum squared resid 8.285993 Schwarz criterion 3.025529
Log likelihood -18.55865 Hannan-Quinn criter. 2.769662
F-statistic 465.3617 Durbin-Watson stat 1.553294
Prob(F-statistic) 0.000000
①与预期不相符。
②评价:
1) 可决系数为 0.994869 ,数据相当大,可以认为拟合程度很好。
2) F 检验, F=465.3617>F (5.12 )=3,89 ,回归方程显著
3) T 检验, X1,X2,X3,X4,X5 ,X6 系数对应的 t 值分别为: 1.254330 ,0.490501 ,-1.005377 ,
0.951671 ,0.476621 ,均小于 t(12)=2.179 ,所以所得系数都是不显著的。
(3)根据 Eviews 分析得到数据如下:
Dependent Variable: Y
Method: Least Squares
Date: 12/03/14 Time: 11:12
Sample: 1994 2011
Included observations: 18
Variable Coefficient Std. Error t-Statistic Prob.
X5 0.001032 2.20E-05 46.79946 0.0000
X6 -0.054965 0.031184 -1.762581 0.0983
C 4.205481 3.335602 1.260786 0.2266
R-squared 0.993601 Mean dependent var 12.76667
Adjusted R-squared 0.992748 S.D. dependent var 9.746631
S.E. of regression 0.830018 Akaike info criterion 2.616274
Sum squared resid 10.33396 Schwarz criterion 2.764669
Log likelihood -20.54646 Hannan-Quinn criter. 2.636736
F-statistic 1164.567 Durbin-Watson stat 1.341880
Prob(F-statistic) 0.000000
①得到模型的方程为:
Y=0.001032 X 5-0.054965 X 6+4.205481
②评价:
1) 可决系数为 0.993601 ,数据相当大,可以认为拟合程度很好。
2) F 检验, F=1164.567>F (5.12 )=3,89 ,回归方程显著
3) T 检验, X5 系数对应的 t 值为 46.79946 ,大于 t(12)=2.179 ,所以系数是显著的,
即人均 GDP 对年底存款余额有显著影响。 X6 系数对应的 t 值为 -1.762581 ,小于 t
(12 )=2.179 ,所以系数是不显著的。
4.3
(1)根据 Eviews 分析得到数据如下:
Dependent Variable: LNY
Method: Least Squares
Date: 12/05/14 Time: 11:39
Sample: 1985 2011
Included observations: 27
Variable Coefficient Std. Error t-Statistic Prob.
LNGDP 1.338533 0.088610 15.10582 0.0000
LNCPI -0.421791 0.233295 -1.807975 0.0832
C -3.111486 0.463010 -6.720126 0.0000
R-squared 0.988051 Mean dependent var 9.484710
Adjusted R-squared 0.987055 S.D. dependent var 1.425517
S.E. of regression 0.162189 Akaike info criterion -0.695670
Sum squared resid 0.631326 Schwarz criterion -0.551689
Log likelihood 12.39155 Hannan-Quinn criter. -0.652857
F-statistic 992.2582 Durbin-Watson stat 0.522613
Prob(F-statistic) 0.000000
得到的模型方程为:
LNY=1.338533 LNGDP t-0.421791 LNCPI t-3.111486
(2)
① 该模型的可决系数为 0.988051 ,可决系数很高, F 检验值为 992.2582 ,
明显显著。但当 α =0.05 时, t(24) =2.064,LNCPI 的系数不显著,可能存在多重共线性。
②得到相关系数矩阵如下:
LNY LNGDP LNCPI
LNY 1.000000 0.993189 0.935116
LNGDP 0.993189 1.000000 0.953740
LNCPI 0.935116 0.953740 1.000000
LNGDP , LNCPI 之间的相关系数很高,证实确实存在多重共线性。
(3)由 Eviews 得:
a)
Dependent Variable: LNY
Method: Least Squares
Date: 12/03/14 Time: 14:41
Sample: 1985 2011
Included observations: 27
Variable Coefficient Std. Error t-Statistic Prob.
LNGDP 1.185739 0.027822 42.61933 0.0000
C -3.750670 0.312255 -12.01156 0.0000
R-squared 0.986423 Mean dependent var 9.484710
Adjusted R-squared 0.985880 S.D. dependent var 1.425517
S.E. of regression 0.169389 Akaike info criterion -0.642056
Sum squared resid 0.717312 Schwarz criterion -0.546068
Log likelihood 10.66776 Hannan-Quinn criter. -0.613514
F-statistic 1816.407 Durbin-Watson stat 0.471111
Prob(F-statistic) 0.000000
b)
Dependent Variable: LNY
Method: Least Squares
Date: 12/03/14 Time: 14:41
Sample: 1985 2011
Included observations: 27
Variable Coefficient Std. Error t-Statistic Prob.
LNCPI 2.939295 0.222756 13.19511 0.0000
C -6.854535 1.242243 -5.517871 0.0000
R-squared 0.874442 Mean dependent var 9.484710
Adjusted R-squared 0.869419 S.D. dependent var 1.425517
S.E. of regression 0.515124 Akaike info criterion 1.582368
Sum squared resid 6.633810 Schwarz criterion 1.678356
Log likelihood -19.36196 Hannan-Quinn criter. 1.610910
F-statistic 174.1108 Durbin-Watson stat 0.137042
Prob(F-statistic) 0.000000
c)
Dependent Variable: LNGDP
Method: Least Squares
Date: 12/05/14 Time: 11:11
Sample: 1985 2011
Included observations: 27
Variable Coefficient Std. Error t-Statistic Prob.
LNCPI 2.511022 0.158302 15.86227 0.0000
C -2.796381 0.882798 -3.167634 0.0040
R-squared 0.909621 Mean dependent var 11.16214
Adjusted R-squared 0.906005 S.D. dependent var 1.194029
S.E. of regression 0.366072 Akaike info criterion 0.899213
Sum squared resid 3.350216 Schwarz criterion 0.995201
Log likelihood -10.13938 Hannan-Quinn criter. 0.927755
F-statistic 251.6117 Durbin-Watson stat 0.099623
Prob(F-statistic) 0.000000
①得到的回归方程分别为
1)LNY=1.185739 LNGDP t-3.750670
2)LNY=2.939295 LNCPI t-6.854535
3)LNGDP t=2.511022 LNCPI t-2.796381
②对多重共线性的认识:
单方程拟合效果都很好,回归系数显著,判定系数较高, GDP和 CPI对进口的显著的单一影
响,在这两个变量同时引入模型时影响方向发生了改变, 这只有通过相关系数的分析才能发
现。
(4)建议:如果仅仅是作预测,可以不在意这种多重共线性,但如果是进行结构分析,还
是应该引起注意的。
4.4
(1)按照
设计
领导形象设计圆作业设计ao工艺污水处理厂设计附属工程施工组织设计清扫机器人结构设计
的理论模型,由 Eviews 分析得:
Dependent Variable: CZSR
Method: Least Squares
Date: 12/03/14 Time: 11:40
Sample: 1985 2011
Included observations: 27
Variable Coefficient Std. Error t-Statistic Prob.
CZZC 0.090114 0.044367 2.031129 0.0540
GDP -0.025334 0.005069 -4.998036 0.0000
SSZE 1.176894 0.062162 18.93271 0.0000
C -221.8540 130.6532 -1.698038 0.1030
R-squared 0.999857 Mean dependent var 22572.56
Adjusted R-squared 0.999838 S.D. dependent var 27739.49
S.E. of regression 353.0540 Akaike info criterion 14.70707
Sum squared resid 2866884. Schwarz criterion 14.89905
Log likelihood -194.5455 Hannan-Quinn criter. 14.76416
F-statistic 53493.93 Durbin-Watson stat 1.458128
Prob(F-statistic) 0.000000
从回归结果可见,可决系数为 0.999857 ,校正的可决系数为 0.999838 ,模型拟合的很好。
F 的统计量为 53493.93 ,说明在 α=0.05, 水平下,回归方程回归方程整体上是显著的。但
是 t 检验结果表明,国内生产总值对财政收入的影响显著,但回归系数的符号为负,与实际
不符合。由此可得知,该方程可能存在多重共线性。
(2)得到相关系数矩阵如下:
CZSR CZZC GDP SSZE
CZSR 1.000000 0.998729 0.992838 0.999832
CZZC 0.998729 1.000000 0.992536 0.998575
GDP 0.992838 0.992536 1.000000 0.994370
SSZE 0.999832 0.998575 0.994370 1.000000
由上表可知, CZZC 与 GDP , CZZC 与 SSZE ,GDP 与 SSZE 之间的相关系数都非常高,
说明确实存在多重共线性。
(3)做辅助回归
被解释变量 可决系数 方差扩大因子
CZZC 0.997168 353
GDP 0.988833 90
SSZE 0.997862 468
方差扩大因子均大于 10,存在严重多重共线性。并且通过以上分析,两两被解释变量之间
相关性都很高。
(4)解决方式:分别作出财政收入与财政支出、国内生产总值、税收总额之间的一元回归。
5.2
(1)
①用图形法检验
绘制 e2 的散点图,用 Eviews 分析如下:
0
5,000
10,000
15,000
20,000
25,000
30,000
1,000 1,500 2,000 2,500 3,000 3,500 4,000
X
E
2
由上图可知,模型可能存在异方差,
② Goldfeld-Quanadt 检验
1)定义区间为 1-7 时,由软件分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/10/14 Time: 14:52
Sample: 1 7
Included observations: 7
Variable Coefficient Std. Error t-Statistic Prob.
T 35.20664 4.901492 7.182843 0.0020
X 0.109949 0.061965 1.774380 0.1507
C 77.12588 82.32844 0.936807 0.4019
R-squared 0.943099 Mean dependent var 565.6857
Adjusted R-squared 0.914649 S.D. dependent var 108.2755
S.E. of regression 31.63265 Akaike info criterion 10.04378
Sum squared resid 4002.499 Schwarz criterion 10.02060
Log likelihood -32.15324 Hannan-Quinn criter. 9.757267
F-statistic 33.14880 Durbin-Watson stat 1.426262
Prob(F-statistic) 0.003238
得∑ e1i
2
=4002.499
2)定义区间为 12-18 时,由软件分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/10/14 Time: 13:50
Sample: 12 18
Included observations: 7
Variable Coefficient Std. Error t-Statistic Prob.
T 52.40588 6.923378 7.569409 0.0016
X 0.068689 0.053763 1.277635 0.2705
C -8.789265 79.92542 -0.109968 0.9177
R-squared 0.984688 Mean dependent var 887.6143
Adjusted R-squared 0.977032 S.D. dependent var 274.4148
S.E. of regression 41.58810 Akaike info criterion 10.59103
Sum squared resid 6918.280 Schwarz criterion 10.56785
Log likelihood -34.06861 Hannan-Quinn criter. 10.30451
F-statistic 128.6166 Durbin-Watson stat 2.390329
Prob(F-statistic) 0.000234
得∑ e2i
2
=6918.280
3)根据 Goldfeld-Quanadt 检验, F 统计量为:
F=∑e2i
2 / ∑e1i2 =6918.280/4002.499=1.7285
在α =0.05 水平下,分子分母的自由度均为 4,查分布表得临界值 F 0.05( 4,4)=6.39 ,因为
F=1.7285< F 0.05( 4,4)=6.39 ,所以接受原假设,此检验表明模型不存在异方差。
(2)存在异方差,估计参数的方法:
①可以对模型进行变换
②使用加权最小二乘法进行计算,得出模型方程,并对其进行相关检验
③对模型进行对数变换,进行分析
(3)评价:
3.3 所得结论是可以相信的,随机扰动项之间不存在异方差。回归方程是显著的。
5.3
(1)由 Eviews 软件分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/10/14 Time: 16:00
Sample: 1 31
Included observations: 31
Variable Coefficient Std. Error t-Statistic Prob.
X 1.244281 0.079032 15.74411 0.0000
C 242.4488 291.1940 0.832602 0.4119
R-squared 0.895260 Mean dependent var 4443.526
Adjusted R-squared 0.891649 S.D. dependent var 1972.072
S.E. of regression 649.1426 Akaike info criterion 15.85152
Sum squared resid 12220196 Schwarz criterion 15.94404
Log likelihood -243.6986 Hannan-Quinn criter. 15.88168
F-statistic 247.8769 Durbin-Watson stat 1.078581
Prob(F-statistic) 0.000000
由上表可知, 2007 年我国农村居民家庭人均消费支出( x)对人均纯收入( y)的模型为:
Y=1.244281X+242.4488
(2)
①由图形法检验
0
1,000,000
2,000,000
3,000,000
4,000,000
5,000,000
6,000,000
0 2,000 4,000 6,000 8,000 10,000
X
E2
由上图可知,模型可能存在异方差。
②Goldfeld-Quanadt 检验
1)定义区间为 1-12 时,由软件分析得:
Dependent Variable: Y1
Method: Least Squares
Date: 12/10/14 Time: 11:34
Sample: 1 12
Included observations: 12
Variable Coefficient Std. Error t-Statistic Prob.
X1 1.485296 0.500386 2.968297 0.0141
C -550.5492 1220.063 -0.451247 0.6614
R-squared 0.468390 Mean dependent var 3052.950
Adjusted R-squared 0.415229 S.D. dependent var 550.5148
S.E. of regression 420.9803 Akaike info criterion 15.07406
Sum squared resid 1772245. Schwarz criterion 15.15488
Log likelihood -88.44437 Hannan-Quinn criter. 15.04414
F-statistic 8.810789 Durbin-Watson stat 2.354167
Prob(F-statistic) 0.014087
得∑ e1i
2
=1772245.
2)定义区间为 20-31 时,由软件分析得:
Dependent Variable: Y1
Method: Least Squares
Date: 12/10/14 Time: 16:36
Sample: 20 31
Included observations: 12
Variable Coefficient Std. Error t-Statistic Prob.
X1 1.086940 0.148863 7.301623 0.0000
C 1173.307 733.2520 1.600141 0.1407
R-squared 0.842056 Mean dependent var 6188.329
Adjusted R-squared 0.826262 S.D. dependent var 2133.692
S.E. of regression 889.3633 Akaike info criterion 16.56990
Sum squared resid 7909670. Schwarz criterion 16.65072
Log likelihood -97.41940 Hannan-Quinn criter. 16.53998
F-statistic 53.31370 Durbin-Watson stat 2.339767
Prob(F-statistic) 0.000026
得∑ e2i
2
=7909670.
3)根据 Goldfeld-Quanadt 检验, F 统计量为:
F=∑e2i2 / ∑e1i2 =7909670./ 1772245=4.4631
在α =0.05 水平下,分子分母的自由度均为 10,查分布表得临界值 F 0.05( 10,10 )=2.98 ,
因为 F=4.4631> F 0.05 (10,10 )=2.98 ,所以拒绝原假设,此检验表明模型存在异方差。
(3)
1)采用 WLS 法估计过程中,
①用权数 w1=1/X, 建立回归得:
Dependent Variable: Y
Method: Least Squares
Date: 12/09/14 Time: 11:13
Sample: 1 31
Included observations: 31
Weighting series: W1
Variable Coefficient Std. Error t-Statistic Prob.
X 1.425859 0.119104 11.97157 0.0000
C -334.8131 344.3523 -0.972298 0.3389
Weighted Statistics
R-squared 0.831707 Mean dependent var 3946.082
Adjusted R-squared 0.825904 S.D. dependent var 536.1907
S.E. of regression 536.6796 Akaike info criterion 15.47102
Sum squared resid 8352726. Schwarz criterion 15.56354
Log likelihood -237.8008 Hannan-Quinn criter. 15.50118
F-statistic 143.3184 Durbin-Watson stat 1.369081
Prob(F-statistic) 0.000000
Unweighted Statistics
R-squared 0.875855 Mean dependent var 4443.526
Adjusted R-squared 0.871574 S.D. dependent var 1972.072
S.E. of regression 706.7236 Sum squared resid 14484289
Durbin-Watson stat 1.532908
对此模型进行 White 检验得:
Heteroskedasticity Test: White
F-statistic 0.299395 Prob. F(2,28) 0.7436
Obs*R-squared 0.649065 Prob. Chi-Square(2) 0.7229
Scaled explained SS 1.798067 Prob. Chi-Square(2) 0.4070
Test Equation:
Dependent Variable: WGT_RESID^2
Method: Least Squares
Date: 12/10/14 Time: 21:13
Sample: 1 31
Included observations: 31
Collinear test regressors dropped from specification
Variable Coefficient Std. Error t-Statistic Prob.
C 61927.89 1045682. 0.059222 0.9532
WGT^2 -593927.9 1173622. -0.506064 0.6168
X*WGT^2 282.4407 747.9780 0.377606 0.7086
R-squared 0.020938 Mean dependent var 269442.8
Adjusted R-squared -0.048995 S.D. dependent var 689166.5
S.E. of regression 705847.6 Akaike info criterion 29.86395
Sum squared resid 1.40E+13 Schwarz criterion 30.00273
Log likelihood -459.8913 Hannan-Quinn criter. 29.90919
F-statistic 0.299395 Durbin-Watson stat 1.922336
Prob(F-statistic) 0.743610
从上可知, nR 2=0.649065 ,比较计算的 统计量的临界值,因为 nR 2=0.649065< 0.05
(2)=5.9915 ,所以接受原假设,该模型消除了异方差。
估计结果为:
Y=1.425859X-334.8131
t=( 11.97157 )( -0.972298 )
R2=0.875855 F=143.3184 DW=1.369081
②用权数 w2=1/x 2,用回归分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/09/14 Time: 21:08
Sample: 1 31
Included observations: 31
Weighting series: W2
Variable Coefficient Std. Error t-Statistic Prob.
X 1.557040 0.145392 10.70922 0.0000
C -693.1946 376.4760 -1.841272 0.0758
Weighted Statistics
R-squared 0.798173 Mean dependent var 3635.028
Adjusted R-squared 0.791214 S.D. dependent var 1029.830
S.E. of regression 466.8513 Akaike info criterion 15.19224
Sum squared resid 6320554. Schwarz criterion 15.28475
Log likelihood -233.4797 Hannan-Quinn criter. 15.22240
F-statistic 114.6875 Durbin-Watson stat 1.562975
Prob(F-statistic) 0.000000
Unweighted Statistics
R-squared 0.834850 Mean dependent var 4443.526
Adjusted R-squared 0.829156 S.D. dependent var 1972.072
S.E. of regression 815.1229 Sum squared resid 19268334
Durbin-Watson stat 1.678365
对此模型进行 White 检验得:
Heteroskedasticity Test: White
F-statistic 0.299790 Prob. F(3,27) 0.8252
Obs*R-squared 0.999322 Prob. Chi-Square(3) 0.8014
Scaled explained SS 1.789507 Prob. Chi-Square(3) 0.6172
Test Equation:
Dependent Variable: WGT_RESID^2
Method: Least Squares
Date: 12/10/14 Time: 21:29
Sample: 1 31
Included observations: 31
Variable Coefficient Std. Error t-Statistic Prob.
C -111661.8 549855.7 -0.203075 0.8406
WGT^2 426220.2 2240181. 0.190262 0.8505
X^2*WGT^2 0.194888 0.516395 0.377402 0.7088
X*WGT^2 -583.2151 2082.820 -0.280012 0.7816
R-squared 0.032236 Mean dependent var 203888.8
Adjusted R-squared -0.075293 S.D. dependent var 419282.0
S.E. of regression 434780.1 Akaike info criterion 28.92298
Sum squared resid 5.10E+12 Schwarz criterion 29.10801
Log likelihood -444.3062 Hannan-Quinn criter. 28.98330
F-statistic 0.299790 Durbin-Watson stat 1.835854
Prob(F-statistic) 0.825233
从上可知, nR 2=0.999322 ,比较计算的 统计量的临界值,因为 nR 2=0.999322< 0.05
(2)=5.9915 ,所以接受原假设,该模型消除了异方差。
估计结果为:
Y=1.557040X-693.1946
t=( 10.70922 )( -1.841272 )
R2=0.798173 F=114.6875 DW=1.562975
③用权数 w3=1/sqr (x),用回归分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/09/14 Time: 21:35
Sample: 1 31
Included observations: 31
Weighting series: W3
Variable Coefficient Std. Error t-Statistic Prob.
X 1.330130 0.098345 13.52507 0.0000
C -47.40242 313.1154 -0.151390 0.8807
Weighted Statistics
R-squared 0.863161 Mean dependent var 4164.118
Adjusted R-squared 0.858442 S.D. dependent var 991.2079
S.E. of regression 586.9555 Akaike info criterion 15.65012
Sum squared resid 9990985. Schwarz criterion 15.74263
Log likelihood -240.5768 Hannan-Quinn criter. 15.68027
F-statistic 182.9276 Durbin-Watson stat 1.237664
Prob(F-statistic) 0.000000
Unweighted Statistics
R-squared 0.890999 Mean dependent var 4443.526
Adjusted R-squared 0.887240 S.D. dependent var 1972.072
S.E. of regression 662.2171 Sum squared resid 12717412
Durbin-Watson stat 1.314859
对此模型进行 White 检验得:
Heteroskedasticity Test: White
F-statistic 0.423886 Prob. F(2,28) 0.6586
Obs*R-squared 0.911022 Prob. Chi-Square(2) 0.6341
Scaled explained SS 2.768332 Prob. Chi-Square(2) 0.2505
Test Equation:
Dependent Variable: WGT_RESID^2
Method: Least Squares
Date: 12/09/14 Time: 20:36
Sample: 1 31
Included observations: 31
Collinear test regressors dropped from specification
Variable Coefficient Std. Error t-Statistic Prob.
C 1212308. 2141958. 0.565981 0.5759
WGT^2 -715673.0 1301839. -0.549740 0.5869
X^2*WGT^2 -0.015194 0.082276 -0.184677 0.8548
R-squared 0.029388 Mean dependent var 322289.8
Adjusted R-squared -0.039942 S.D. dependent var 863356.7
S.E. of regression 880429.8 Akaike info criterion 30.30597
Sum squared resid 2.17E+13 Schwarz criterion 30.44475
Log likelihood -466.7426 Hannan-Quinn criter. 30.35121
F-statistic 0.423886 Durbin-Watson stat 1.887426
Prob(F-statistic) 0.658628
从上可知, nR 2=0.911022 ,比较计算的 统计量的临界值,因为 nR 2=0.911022< 0.05
(2)=5.9915 ,所以接受原假设,该模型消除了异方差。
估计结果为:
Y=1.330130X-47.40242
t=( 13.52507 )( -0.151390 )
R2=0.863161 F=182.9276 DW=1.237664
经过检验发现,用权数 w1 的效果最好,所以综上可知,即修改后的结果为:
Y=1.425859X-334.8131
t=( 11.97157 )( -0.972298 )
R2=0.875855 F=143.3184 DW=1.369081
5.6
(1)
a)用 Eviews 模型分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/10/14 Time: 20:16
Sample: 1978 2011
Included observations: 34
Variable Coefficient Std. Error t-Statistic Prob.
X 0.746241 0.019120 39.03027 0.0000
C 92.55422 42.80529 2.162215 0.0382
R-squared 0.979426 Mean dependent var 1295.802
Adjusted R-squared 0.978783 S.D. dependent var 1188.791
S.E. of regression 173.1597 Akaike info criterion 13.20333
Sum squared resid 959497.2 Schwarz criterion 13.29311
Log likelihood -222.4566 Hannan-Quinn criter. 13.23395
F-statistic 1523.362 Durbin-Watson stat 1.534491
Prob(F-statistic) 0.000000
得回归模型为:
Y=0.746241 X+92.55422
b)检验是否存在异方差:
①用 Goldfeld-Quanadt 检验如下:
1)当定义区间为 1-13 时,由软件分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/11/14 Time: 11:47
Sample: 1 13
Included observations: 13
Variable Coefficient Std. Error t-Statistic Prob.
X 0.967839 0.026879 36.00771 0.0000
C -18.86861 8.963780 -2.104984 0.0591
R-squared 0.991587 Mean dependent var 280.1377
Adjusted R-squared 0.990823 S.D. dependent var 127.0409
S.E. of regression 12.17039 Akaike info criterion 7.976527
Sum squared resid 1629.301 Schwarz criterion 8.063442
Log likelihood -49.84742 Hannan-Quinn criter. 7.958662
F-statistic 1296.555 Durbin-Watson stat 1.071505
Prob(F-statistic) 0.000000
得∑ e1i
2
=1629.301
2)当定义区间为 1-13 时,由软件分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/11/14 Time: 12:21
Sample: 22 34
Included observations: 13
Variable Coefficient Std. Error t-Statistic Prob.
X 0.719567 0.058312 12.33998 0.0000
C 179.3950 202.8764 0.884258 0.3955
R-squared 0.932629 Mean dependent var 2496.127
Adjusted R-squared 0.926504 S.D. dependent var 1022.591
S.E. of regression 277.2250 Akaike info criterion 14.22817
Sum squared resid 845390.4 Schwarz criterion 14.31509
Log likelihood -90.48313 Hannan-Quinn criter. 14.21031
F-statistic 152.2752 Durbin-Watson stat 1.658418
Prob(F-statistic) 0.000000
得∑ e2i
2
=845390.4
3)根据 Goldfeld-Quanadt 检验, F 统计量为:
F=∑e2i
2 / ∑e1i2 =845390.4/ 1629.301=518.8669
在α =0.05 水平下,分子分母的自由度均为 11,查分布表得临界值 F 0.05( 11,11 )=4.47 ,
因为 F=518.8669> F 0.05(11,11 )=4.47 ,所以拒绝原假设,此检验表明模型存在异方差。
②White 检验
用 EViews 软件分析得:
Heteroskedasticity Test: White
F-statistic 10.36759 Prob. F(2,31) 0.0004
Obs*R-squared 13.62701 Prob. Chi-Square(2) 0.0011
Scaled explained SS 76.13635 Prob. Chi-Square(2) 0.0000
Test Equation:
Dependent Variable: RESID^2
Method: Least Squares
Date: 12/11/14 Time: 12:56
Sample: 1 34
Included observations: 34
Variable Coefficient Std. Error t-Statistic Prob.
C 11581.11 26117.11 0.443430 0.6605
X -27.69901 27.86540 -0.994029 0.3279
X^2 0.012230 0.005156 2.371861 0.0241
R-squared 0.400795 Mean dependent var 28220.51
Adjusted R-squared 0.362136 S.D. dependent var 101738.9
S.E. of regression 81255.15 Akaike info criterion 25.53267
Sum squared resid 2.05E+11 Schwarz criterion 25.66735
Log likelihood -431.0554 Hannan-Quinn criter. 25.57860
F-statistic 10.36759 Durbin-Watson stat 3.021651
Prob(F-statistic) 0.000357
从 上 图 中 可 以 看 出 , nR 2=13.62701 , 比 较 计 算 的 统 计 量 的 临 界 值 , 因 为
nR 2=13.62701> 0.05 (2)=5.9915 ,所以拒绝原假设,不拒绝备择假设,表明模型存在
异方差。
用以上两种方法,可以检验模型是存在异方差的。
c)修正模型
1)用加权二乘法修正异方差现象步骤如下:
①当权数 w1=1/x 时,用软件分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/11/14 Time: 13:22
Sample: 1 34
Included observations: 34
Weighting series: W1
Variable Coefficient Std. Error t-Statistic Prob.
X 0.821013 0.016866 48.67993 0.0000
C 17.69318 6.283256 2.815926 0.0083
Weighted Statistics
R-squared 0.986676 Mean dependent var 457.8505
Adjusted R-squared 0.986260 S.D. dependent var 41.70384
S.E. of regression 37.91285 Akaike info criterion 10.16548
Sum squared resid 45996.29 Schwarz criterion 10.25527
Log likelihood -170.8132 Hannan-Quinn criter. 10.19610
F-statistic 2369.735 Durbin-Watson stat 0.605852
Prob(F-statistic) 0.000000
Unweighted Statistics
R-squared 0.968070 Mean dependent var 1295.802
Adjusted R-squared 0.967072 S.D. dependent var 1188.791
S.E. of regression 215.7175 Sum squared resid 1489089.
Durbin-Watson stat 1.079107
得方程模型为:
Y=0.821013X-17.69318
t=( 48.67993 )( 2.815926 )
R2=0.986676 F=2369.735 DW=0.605852
对此模型进行 White 检验如下:
Heteroskedasticity Test: White
F-statistic 1.348072 Prob. F(2,31) 0.2745
Obs*R-squared 2.720457 Prob. Chi-Square(2) 0.2566
Scaled explained SS 1.221901 Prob. Chi-Square(2) 0.5428
Test Equation:
Dependent Variable: WGT_RESID^2
Method: Least Squares
Date: 12/11/14 Time: 11:20
Sample: 1 34
Included observations: 34
Collinear test regressors dropped from specification
Variable Coefficient Std. Error t-Statistic Prob.
C 1678.870 416.5417 4.030498 0.0003
WGT^2 -32.13071 187.6175 -0.171257 0.8651
X*WGT^2 -0.484040 1.279449 -0.378319 0.7078
R-squared 0.080013 Mean dependent var 1352.832
Adjusted R-squared 0.020659 S.D. dependent var 1382.825
S.E. of regression 1368.467 Akaike info criterion 17.36487
Sum squared resid 58053732 Schwarz criterion 17.49955
Log likelihood -292.2027 Hannan-Quinn criter. 17.41080
F-statistic 1.348072 Durbin-Watson stat 1.199640
Prob(F-statistic) 0.274545
从上图中可以看出, nR 2=2.720457 ,比较计算的 统计量的临界值,
因为 nR 2=2.720457< 0.05 (2) =5.9915 ,所以接受原假设,即该模型消除了异方差的影
响。
②当权数 w2=1/x 2 时,用软件分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/11/14 Time: 13:27
Sample: 1 34
Included observations: 34
Weighting series: W2
Variable Coefficient Std. Error t-Statistic Prob.
X 0.852193 0.020150 42.29335 0.0000
C 8.890886 3.604301 2.466744 0.0192
Weighted Statistics
R-squared 0.982425 Mean dependent var 230.2433
Adjusted R-squared 0.981875 S.D. dependent var 247.1718
S.E. of regression 16.20273 Akaike info criterion 8.465259
Sum squared resid 8400.912 Schwarz criterion 8.555045
Log likelihood -141.9094 Hannan-Quinn criter. 8.495879
F-statistic 1788.728 Durbin-Watson stat 0.604647
Prob(F-statistic) 0.000000
Unweighted Statistics
R-squared 0.954142 Mean dependent var 1295.802
Adjusted R-squared 0.952709 S.D. dependent var 1188.791
S.E. of regression 258.5207 Sum squared resid 2138654.
Durbin-Watson stat 0.781788
得方程模型为:
Y=0.852193X+8.890886
t=(42.29335 )( 2.466744 )
R2=0.982425 F=1788.728 DW=0.604647
用 White 检验模型得:
Heteroskedasticity Test: White
F-statistic 7.462185 Prob. F(3,30) 0.0007
Obs*R-squared 14.52935 Prob. Chi-Square(3) 0.0023
Scaled explained SS 19.40139 Prob. Chi-Square(3) 0.0002
Test Equation:
Dependent Variable: WGT_RESID^2
Method: Least Squares
Date: 12/11/14 Time: 11:19
Sample: 1 34
Included observations: 34
Variable Coefficient Std. Error t-Statistic Prob.
C -7.684700 85.76169 -0.089605 0.9292
WGT^2 64.20016 96.11160 0.667975 0.5093
X^2*WGT^2 0.006306 0.003431 1.838317 0.0759
X*WGT^2 -1.247222 1.163558 -1.071903 0.2923
R-squared 0.427334 Mean dependent var 247.0857
Adjusted R-squared 0.370067 S.D. dependent var 435.4791
S.E. of regression 345.6323 Akaike info criterion 14.63876
Sum squared resid 3583851. Schwarz criterion 14.81833
Log likelihood -244.8589 Hannan-Quinn criter. 14.70000
F-statistic 7.462185 Durbin-Watson stat 1.586012
Prob(F-statistic) 0.000712
从 上 图 中 可 以 看 出 , nR 2=14.52935 , 比 较 计 算 的 统 计 量 的 临 界 值 , 因 为
nR 2=14.52935> 0.05 (2)=5.9915 ,所以拒绝原假设,不拒绝备择假设,表明模型存在
异方差。此模型并未消除异方差。
③当权数 w3=1/sqr(x) 时,用软件分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/11/14 Time: 13:21
Sample: 1 34
Included observations: 34
Weighting series: W3
Variable Coefficient Std. Error t-Statistic Prob.
X 0.778551 0.015677 49.66347 0.0000
C 40.45770 14.57528 2.775775 0.0091
Weighted Statistics
R-squared 0.987192 Mean dependent var 776.3266
Adjusted R-squared 0.986792 S.D. dependent var 367.3152
S.E. of regression 79.19828 Akaike info criterion 11.63881
Sum squared resid 200715.8 Schwarz criterion 11.72859
Log likelihood -195.8597 Hannan-Quinn criter. 11.66943
F-statistic 2466.460 Durbin-Watson stat 1.178340
Prob(F-statistic) 0.000000
Unweighted Statistics
R-squared 0.977590 Mean dependent var 1295.802
Adjusted R-squared 0.976890 S.D. dependent var 1188.791
S.E. of regression 180.7210 Sum squared resid 1045123.
Durbin-Watson stat 1.460832
得方程模型为:
Y=0.778551X+40.45770
t=(49.66347 )( 2.775775 )
R2=0.986792 F=2466.460 DW=1.178340
对所得模型进行 White 检验:
Heteroskedasticity Test: White
F-statistic 8.158958 Prob. F(2,31) 0.0014
Obs*R-squared 11.72514 Prob. Chi-Square(2) 0.0028
Scaled explained SS 28.08353 Prob. Chi-Square(2) 0.0000
Test Equation:
Dependent Variable: WGT_RESID^2
Method: Least Squares
Date: 12/10/14 Time: 13:23
Sample: 1 34
Included observations: 34
Collinear test regressors dropped from specification
Variable Coefficient Std. Error t-Statistic Prob.
C -7585.186 5311.263 -1.428132 0.1633
WGT^2 2468.369 1996.041 1.236632 0.2255
X^2*WGT^2 0.009139 0.002481 3.684177 0.0009
R-squared 0.344857 Mean dependent var 5903.405
Adjusted R-squared 0.302590 S.D. dependent var 13934.64
S.E. of regression 11636.97 Akaike info criterion 21.64586
Sum squared resid 4.20E+09 Schwarz criterion 21.78054
Log likelihood -364.9796 Hannan-Quinn criter. 21.69179
F-statistic 8.158958 Durbin-Watson stat 2.344068
Prob(F-statistic) 0.001423
从 上 图 中 可 以 看 出 , nR 2=11.72514 , 比 较 计 算 的 统 计 量 的 临 界 值 , 因 为
nR 2=11.72514> 0.05 (2)=5.9915 ,所以拒绝原假设,不拒绝备择假设,表明模型存在
异方差。此模型并未消除异方差。
综上所述,用加权二乘法 w1 的效果最好,所以模型为:
得方程模型为:
Y=0.821013X-17.69318
t=( 48.67993 )( 2.815926 )
R2=0.986676 F=2369.735 DW=0.605852
2)用对数模型法
用软件分析得:
Dependent Variable: LNY
Method: Least Squares
Date: 12/11/14 Time: 09:54
Sample: 1 34
Included observations: 34
Variable Coefficient Std. Error t-Statistic Prob.
LNX 0.946887 0.011228 84.33549 0.0000
C 0.201861 0.077905 2.591100 0.0143
R-squared 0.995521 Mean dependent var 6.687779
Adjusted R-squared 0.995381 S.D. dependent var 1.067124
S.E. of regression 0.072525 Akaike info criterion -2.352753
Sum squared resid 0.168315 Schwarz criterion -2.262967
Log likelihood 41.99680 Hannan-Quinn criter. -2.322134
F-statistic 7112.475 Durbin-Watson stat 0.812150
Prob(F-statistic) 0.000000
得到模型为:
LnY=0.946887 LNX+0.201861
对此模型进行 White 检验得:
Heteroskedasticity Test: White
F-statistic 1.003964 Prob. F(2,31) 0.3780
Obs*R-squared 2.068278 Prob. Chi-Square(2) 0.3555
Scaled explained SS 1.469638 Prob. Chi-Square(2) 0.4796
Test Equation:
Dependent Variable: RESID^2
Method: Least Squares
Date: 12/11/14 Time: 09:55
Sample: 1 34
Included observations: 34
Variable Coefficient Std. Error t-Statistic Prob.
C 0.039547 0.046759 0.845753 0.4042
LNX -0.011601 0.014012 -0.827969 0.4140
LNX^2 0.000932 0.001028 0.906774 0.3715
R-squared 0.060832 Mean dependent var 0.004950
Adjusted R-squared 0.000240 S.D. dependent var 0.006365
S.E. of regression 0.006364 Akaike info criterion -7.192271
Sum squared resid 0.001255 Schwarz criterion -7.057592
Log likelihood 125.2686 Hannan-Quinn criter. -7.146342
F-statistic 1.003964 Durbin-Watson stat 2.022904
Prob(F-statistic) 0.378027
从 上 图 中 可 以 看 出 , nR 2=2.068278 , 比 较 计 算 的 统 计 量 的 临 界 值 , 因 为
nR 2=2.068278< 0.05 (2)=5.9915 ,所以接受原假设,此模型消除了异方差。
综合两种方法,改进后的模型最好为:
LnY=0.946887 LNX+0.201861
(2)
1)考虑价格因素,首先用软件三者关系进行分析如下:
Dependent Variable: Y
Method: Least Squares
Date: 12/12/14 Time: 19:26
Sample: 1 34
Included observations: 34
Variable Coefficient Std. Error t-Statistic Prob.
X 0.741684 0.019905 37.26095 0.0000
P 0.235025 0.271701 0.865012 0.3937
C 43.41715 71.22946 0.609539 0.5466
R-squared 0.979911 Mean dependent var 1295.802
Adjusted R-squared 0.978615 S.D. dependent var 1188.791
S.E. of regression 173.8449 Akaike info criterion 13.23830
Sum squared resid 936883.7 Schwarz criterion 13.37298
Log likelihood -222.0511 Hannan-Quinn criter. 13.28423
F-statistic 756.0627 Durbin-Watson stat 1.681521
Prob(F-statistic) 0.000000
1)用 Goldfeld-Quanadt 检验如下:
①当样本为 1-13 时,进行回归分析:
Dependent Variable: P
Method: Least Squares
Date: 12/14/14 Time: 19:26
Sample: 1 13
Included observations: 13
Variable Coefficient Std. Error t-Statistic Prob.
X -0.170484 0.203868 -0.836247 0.4225
Y 0.458660 0.209755 2.186646 0.0536
C 59.50496 7.385841 8.056627 0.0000
R-squared 0.956255 Mean dependent var 135.3231
Adjusted R-squared 0.947506 S.D. dependent var 36.95380
S.E. of regression 8.466678 Akaike info criterion 7.309328
Sum squared resid 716.8464 Schwarz criterion 7.439701
Log likelihood -44.51063 Hannan-Quinn criter. 7.282530
F-statistic 109.2993 Durbin-Watson stat 0.637181
Prob(F-statistic) 0.000000
得∑ e1i
2
=716.8464
②当样本为 22-34 时,做回归分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/14/14 Time:20:39
Sample: 22 34
Included observations: 13
Variable Coefficient Std. Error t-Statistic Prob.
X 0.641197 0.092678 6.918569 0.0000
P -1.206222 1.114278 -1.082514 0.3044
C 795.6887 603.8605 1.317670 0.2170
R-squared 0.939696 Mean dependent var 2496.127
Adjusted R-squared 0.927635 S.D. dependent var 1022.591
S.E. of regression 275.0847 Akaike info criterion 14.27121
Sum squared resid 756715.7 Schwarz criterion 14.40158
Log likelihood -89.76286 Hannan-Quinn criter. 14.24441
F-statistic 77.91291 Durbin-Watson stat 1.128778
Prob(F-statistic) 0.000001
得∑ e2i
2
=756715.7
③根据 Goldfeld-Quanadt 检验, F 统计量为:
F=∑e2i2 / ∑e1i2 =756715.7/ 716.8464=1055.6176
在α =0.05 水平下,分子分母的自由度均为 11,查分布表得临界值 F 0.05( 10,10 )=2.98 ,
因为 F=1055.6176> F 0.05 (10,10 )=2.98 ,所以拒绝原假设,此检验表明模型存在异方差。
2)用 White 检验,软件分析结果为:
Heteroskedasticity Test: White
F-statistic 7.312529 Prob. F(5,28) 0.0002
Obs*R-squared 19.25463 Prob. Chi-Square(5) 0.0017
Scaled explained SS 119.3072 Prob. Chi-Square(5) 0.0000
Test Equation:
Dependent Variable: RESID^2
Method: Least Squares
Date: 12/12/14 Time: 19:31
Sample: 1 34
Included observations: 34
Variable Coefficient Std. Error t-Statistic Prob.
C 79541.08 112647.3 0.706107 0.4860
X 209.4964 63.90400 3.278298 0.0028
X^2 -0.024133 0.010712 -2.252841 0.0323
X*P -0.235137 0.106647 -2.204822 0.0358
P -1175.326 1156.253 -1.016495 0.3181
P^2 1.637366 2.600020 0.629751 0.5340
R-squared 0.566313 Mean dependent var 27555.40
Adjusted R-squared 0.488869 S.D. dependent var 107990.9
S.E. of regression 77206.44 Akaike info criterion 25.50514
Sum squared resid 1.67E+11 Schwarz criterion 25.77450
Log likelihood -427.5874 Hannan-Quinn criter. 25.59700
F-statistic 7.312529 Durbin-Watson stat 2.787044
Prob(F-statistic) 0.000171
从 上 图 中 可 以 看 出 , nR 2=19.25463 , 比 较 计 算 的 统 计 量 的 临 界 值 , 因 为
nR 2=19.25463> 0.05(5)=11.0705 ,所以拒绝原假设,不拒绝备择假设,表明模型存在
异方差。
2)修正
①建立对数模型,用软件分析如下:
Dependent Variable: LNY
Method: Least Squares
Date: 12/12/14 Time: 19:24
Sample: 1 34
Included observations: 34
Variable Coefficient Std. Error t-Statistic Prob.
LNX 0.939605 0.013645 68.86088 0.0000
LNP 0.026821 0.028454 0.942609 0.3532
C 0.108230 0.126322 0.856784 0.3981
R-squared 0.995646 Mean dependent var 6.687779
Adjusted R-squared 0.995365 S.D. dependent var 1.067124
S.E. of regression 0.072652 Akaike info criterion -2.322188
Sum squared resid 0.163625 Schwarz criterion -2.187509
Log likelihood 42.47720 Hannan-Quinn criter. -2.276259
F-statistic 3544.292 Durbin-Watson stat 0.930109
Prob(F-statistic) 0.000000
对此模型进行 White 检验:
Heteroskedasticity Test: White
F-statistic 3.523832 Prob. F(5,28) 0.0135
Obs*R-squared 13.13158 Prob. Chi-Square(5) 0.0222
Scaled explained SS 12.14373 Prob. Chi-Square(5) 0.0329
Test Equation:
Dependent Variable: RESID^2
Method: Least Squares
Date: 12/12/14 Time: 19:24
Sample: 1 34
Included observations: 34
Variable Coefficient Std. Error t-Statistic Prob.
C 0.422872 0.273746 1.544759 0.1336
LNX 0.080712 0.031833 2.535502 0.0171
LNX^2 -0.003917 0.003037 -1.289564 0.2078
LNX*LNP -0.004955 0.005136 -0.964765 0.3429
LNP -0.254992 0.129858 -1.963631 0.0596
LNP^2 0.026470 0.012675 2.088390 0.0460
R-squared 0.386223 Mean dependent var 0.004813
Adjusted R-squared 0.276620 S.D. dependent var 0.007286
S.E. of regression 0.006197 Akaike info criterion -7.170690
Sum squared resid 0.001075 Schwarz criterion -6.901332
Log likelihood 127.9017 Hannan-Quinn criter. -7.078831
F-statistic 3.523832 Durbin-Watson stat 2.264261
Prob(F-statistic) 0.013502
从 上 图 中 可 以 看 出 , nR 2=13.13158 , 比 较 计 算 的 统 计 量 的 临 界 值 , 因 为
nR 2=13.13158> 0.05(5)=11.0705 ,所以拒绝原假设,不拒绝备择假设,表明模型存在
异方差,所以此模型没有消除异方差。
②当 w1=1/x 时,用软件分析如下:
Dependent Variable: Y
Method: Least Squares
Date: 12/13/14 Time: 18:49
Sample: 1 34
Included observations: 34
Weighting series: W1
Variable Coefficient Std. Error t-Statistic Prob.
X 0.723218 0.022965 31.49212 0.0000
P 0.719506 0.141085 5.099795 0.0000
C -44.72084 13.11268 -3.410502 0.0018
Weighted Statistics
R-squared 0.992755 Mean dependent var 457.8505
Adjusted R-squared 0.992287 S.D. dependent var 41.70384
S.E. of regression 28.40494 Akaike info criterion 9.615100
Sum squared resid 25012.05 Schwarz criterion 9.749779
Log likelihood -160.4567 Hannan-Quinn criter. 9.661030
F-statistic 2123.843 Durbin-Watson stat 1.298389
Prob(F-statistic) 0.000000
Unweighted Statistics
R-squared 0.977704 Mean dependent var 1295.802
Adjusted R-squared 0.976266 S.D. dependent var 1188.791
S.E. of regression 183.1446 Sum squared resid 1039800.
Durbin-Watson stat 1.740795
所得模型为:
Y=0.723218X+0.719506p-44.72084
对此模型进行 White 检验得:
Heteroskedasticity Test: White
F-statistic 2.088840 Prob. F(5,28) 0.0966
Obs*R-squared 9.236835 Prob. Chi-Square(5) 0.1000
Scaled explained SS 25.50696 Prob. Chi-Square(5) 0.0001
Test Equation:
Dependent Variable: WGT_RESID^2
Method: Least Squares
Date: 12/14/14 Time: 19:57
Sample: 1 34
Included observations: 34
Collinear test regressors dropped from specification
Variable Coefficient Std. Error t-Statistic Prob.
C 3861.793 1068.806 3.613183 0.0012
WGT^2 3260.199 4309.988 0.756429 0.4557
X*WGT^2 13.72241 8.453473 1.623287 0.1157
X*P*WGT^2 -0.151725 0.061588 -2.463567 0.0202
P^2*WGT^2 0.431162 0.278315 1.549186 0.1326
P*WGT^2 -76.13221 73.40636 -1.037134 0.3085
R-squared 0.271672 Mean dependent var 735.6486
Adjusted R-squared 0.141613 S.D. dependent var 1924.655
S.E. of regression 1783.177 Akaike info criterion 17.96897
Sum squared resid 89032169 Schwarz criterion 18.23832
Log likelihood -299.4724 Hannan-Quinn criter. 18.06082
F-statistic 2.088840 Durbin-Watson stat 2.336495
Prob(F-statistic) 0.096616
因为 nR 2=9.236835< 0.05 (5)=11.0705 ,所以接受原假设。该模型不存在异方差,所
以此模型消除了异方差。
③当 w2=1/x 2,用软件分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/15/14 Time: 20:02
Sample: 1 34
Included observations: 34
Weighting series: W2
Variable Coefficient Std. Error t-Statistic Prob.
X 0.639012 0.039216 16.29477 0.0000
P 1.200751 0.206023 5.828234 0.0000
C -81.85973 15.77499 -5.189209 0.0000
Weighted Statistics
R-squared 0.991614 Mean dependent var 230.2433
Adjusted R-squared 0.991073 S.D. dependent var 247.1718
S.E. of regression 11.37136 Akaike info criterion 7.784170
Sum squared resid 4008.543 Schwarz criterion 7.918849
Log likelihood -129.3309 Hannan-Quinn criter. 7.830100
F-statistic 1832.775 Durbin-Watson stat 1.167961
Prob(F-statistic) 0.000000
Unweighted Statistics
R-squared 0.956816 Mean dependent var 1295.802
Adjusted R-squared 0.954030 S.D. dependent var 1188.791
S.E. of regression 254.8849 Sum squared resid 2013955.
Durbin-Watson stat 1.002870
所得模型为:
Y=0.639012X+1.200751p-81.85973
对该模型进行 White 检验得:
Heteroskedasticity Test: White
F-statistic 43.19853 Prob. F(6,27) 0.0000
Obs*R-squared 30.79235 Prob. Chi-Square(6) 0.0000
Scaled explained SS 47.42430 Prob. Chi-Square(6) 0.0000
Test Equation:
Dependent Variable: WGT_RESID^2
Method: Least Squares
Date: 12/14/14 Time: 19:20
Sample: 1 34
Included observations: 34
Variable Coefficient Std. Error t-Statistic Prob.
C 27.51002 20.12556 1.366919 0.1829
WGT^2 -1245.193 837.2352 -1.487268 0.1485
X^2*WGT^2 0.007732 0.005450 1.418649 0.1674
X*WGT^2 7.948582 4.884597 1.627275 0.1153
X*P*WGT^2 -0.111755 0.064061 -1.744525 0.0924
P^2*WGT^2 0.184342 0.164562 1.120199 0.2725
P*WGT^2 -3.127017 23.56724 -0.132685 0.8954
R-squared 0.905657 Mean dependent var 117.8983
Adjusted R-squared 0.884692 S.D. dependent var 230.3570
S.E. of regression 78.22224 Akaike info criterion 11.73823
Sum squared resid 165205.4 Schwarz criterion 12.05248
Log likelihood -192.5498 Hannan-Quinn criter. 11.84539
F-statistic 43.19853 Durbin-Watson stat 1.794799
Prob(F-statistic) 0.000000
因为 nR 2=30.79235> 0.05 (5)=11.0705 ,所以拒绝原假设,不拒绝备择假设,表明模
型存在异方差,所以此模型没有消除异方差。
④当 w3=1/sqr(x) 时,用软件分析得:
Dependent Variable: Y
Method: Least Squares
Date: 12/14/14 Time: 19:06
Sample: 1 34
Included observations: 34
Weighting series: W3
Variable Coefficient Std. Error t-Statistic Prob.
X 0.744661 0.019825 37.56252 0.0000
P 0.451861 0.179971 2.510739 0.0175
C -13.49643 25.37768 -0.531823 0.5986
Weighted Statistics
R-squared 0.989356 Mean dependent var 776.3266
Adjusted R-squared 0.988670 S.D. dependent var 367.3152
S.E. of regression 73.35237 Akaike info criterion 11.51252
Sum squared resid 166797.7 Schwarz criterion 11.64720
Log likelihood -192.7129 Hannan-Quinn criter. 11.55845
F-statistic 1440.783 Durbin-Watson stat 1.599590
Prob(F-statistic) 0.000000
Unweighted Statistics
R-squared 0.979407 Mean dependent var 1295.802
Adjusted R-squared 0.978079 S.D. dependent var 1188.791
S.E. of regression 176.0098 Sum squared resid 960362.6
Durbin-Watson stat 1.761225
所得模型为:
Y=0.744661X+0.451861p-13.49643
对所得模型进行 White 检验得:
Heteroskedasticity Test: White
F-statistic 4.459272 Prob. F(5,28) 0.0041
Obs*R-squared 15.07219 Prob. Chi-Square(5) 0.0101
Scaled explained SS 72.39077 Prob. Chi-Square(5) 0.0000
Test Equation:
Dependent Variable: WGT_RESID^2
Method: Least Squares
Date: 12/14/14 Time: 19:08
Sample: 1 34
Included observations: 34
Collinear test regressors dropped from specification
Variable Coefficient Std. Error t-Statistic Prob.
C 61163.22 27531.93 2.221538 0.0346
WGT^2 28251.98 17350.39 1.628320 0.1147
X^2*WGT^2 -0.001093 0.006624 -0.164950 0.8702
X*P*WGT^2 -0.235836 0.077110 -3.058447 0.0049
P^2*WGT^2 1.236884 0.644872 1.918030 0.0654
P*WGT^2 -503.3080 262.5884 -1.916718 0.0655
R-squared 0.443300 Mean dependent var 4905.814
Adjusted R-squared 0.343889 S.D. dependent var 16926.97
S.E. of regression 13710.96 Akaike info criterion 22.04856
Sum squared resid 5.26E+09 Schwarz criterion 22.31792
Log likelihood -368.8256 Hannan-Quinn criter. 22.14042
F-statistic 4.459272 Durbin-Watson stat 2.450171
Prob(F-statistic) 0.004103
因为 nR 2=15.07219> 0.05 (5)=11.0705 ,所以拒绝原假设,不拒绝备择假设,表明模
型存在异方差,所以此模型没有消除异方差。
综上所述,修改后的模型为:
Y= Y=0.723218X+0.719506p-44.72084
t=(31.49212) (5.099705) (-3.410502)
R2=0.992755 F=2123.843 DW=1.298389
(3) 体会:对于不同的模型,可采取对数模型法或者加权二乘法对具有异方差性的模型进行
改进,从而消除异方差。但对于不同的模型,自由度的不同,可能导致改进的方法不同,所
以要对改进的模型进行进一步的检验才行。
6.1
(1) 建立居民收入 -消费模型,用 Eviews 分析结果如下:
Dependent Variable: Y
Method: Least Squares
Date: 12/20/14 Time: 14:22
Sample: 1 19
Included observations: 19
Variable Coefficient Std. Error t-Statistic Prob.
X 0.690488 0.012877 53.62068 0.0000
C 79.93004 12.39919 6.446390 0.0000
R-squared 0.994122 Mean dependent var 700.2747
Adjusted R-squared 0.993776 S.D. dependent var 246.4491
S.E. of regression 19.44245 Akaike info criterion 8.872095
Sum squared resid 6426.149 Schwarz criterion 8.971510
Log likelihood -82.28490 Hannan-Quinn criter. 8.888920
F-statistic 2875.178 Durbin-Watson stat 0.574663
Prob(F-statistic) 0.000000
所得模型为:
Y=0.690488X+79.93004
Se=(0.012877)(12.39919)
t=(53.62068)(6.446390)
R2=0.994122 F=2875.178 DW=0.574663
(2)
1)检验模型中存在的问题
①做出残差图如下:
-40
-30
-20
-10
0
10
20
30
40
50
2 4 6 8 10 12 14 16 18
Y Residuals
残差的变动有系统模式,连续为正和连续为负,表明残差项存在一阶自相关。
②该回归方程可决系数较高,回归系数均显著。对样本量为 19 ,一个解释变量的模型, 5%
的显著水平,查 DW 统计表可知, dL=1.180 , dU =1.401 ,模型中 DW=0.574663,< dL,显然
模型中有自相关。
③对模型进行 BG 检验,用 Eviews 分析结果如下:
Breusch-Godfrey Serial Correlation LM Test:
F-statistic 4.811108 Prob. F(2,15) 0.0243
Obs*R-squared 7.425088 Prob. Chi-Square(2) 0.0244
Test Equation:
Dependent Variable: RESID
Method: Least Squares
Date: 12/20/14 Time: 15:03
Sample: 1 19
Included observations: 19
Presample missing value lagged residuals set to zero.
Variable Coefficient Std. Error t-Statistic Prob.
X -0.003275 0.010787 -0.303586 0.7656
C 1.929546 10.35593 0.186323 0.8547
RESID(-1) 0.608886 0.292707 2.080189 0.0551
RESID(-2) 0.089988 0.291120 0.309110 0.7615
R-squared 0.390794 Mean dependent var -1.65E-13
Adjusted R-squared 0.268953 S.D. dependent var 18.89466
S.E. of regression 16.15518 Akaike info criterion 8.587023
Sum squared resid 3914.848 Schwarz criterion 8.785852
Log likelihood -77.57671 Hannan-Quinn criter. 8.620672
F-statistic 3.207406 Durbin-Watson stat 1.570723
Prob(F-statistic) 0.053468
如上表显示, LM=TR2=7.425088 ,其 p 值为 0.0244 ,表明存在自相关。
2)对模型进行处理:
①采取广义差分法
a)为估计自相关系数 ρ。对 e t 进行滞后一期的自回归,用 EViews 分析结果如下:
Dependent Variable: E
Method: Least Squares
Date: 12/20/14 Time: 15:04
Sample (adjusted): 2 19
Included observations: 18 after adjustments
Variable Coefficient Std. Error t-Statistic Prob.
E(-1) 0.657352 0.177626 3.700759 0.0018
R-squared 0.440747 Mean dependent var 1.717433
Adjusted R-squared 0.440747 S.D. dependent var 17.85134
S.E. of regression 13.34980 Akaike info criterion 8.074833
Sum squared resid 3029.692 Schwarz criterion 8.124298
Log likelihood -71.67349 Hannan-Quinn criter. 8.081653
Durbin-Watson stat 1.634573
由上可知, ρ=0.657352
b)对原模型进行广义差分回归,用 Eviews 进行分析所得结果如下:
Dependent Variable: Y-0.657352*Y(-1)
Method: Least Squares
Date: 12/20/14 Time: 15:04
Sample (adjusted): 2 19
Included observations: 18 after adjustments
Variable Coefficient Std. Error t-Statistic Prob.
C 35.97761 8.103546 4.439737 0.0004
X-0.657352*X(-1) 0.668695 0.020642 32.39512 0.0000
R-squared 0.984983 Mean dependent var 278.1002
Adjusted R-squared 0.984044 S.D. dependent var 105.1781
S.E. of regression 13.28570 Akaike info criterion 8.115693
Sum squared resid 2824.158 Schwarz criterion 8.214623
Log likelihood -71.04124 Hannan-Quinn criter. 8.129334
F-statistic 1049.444 Durbin-Watson stat 1.830746
Prob(F-statistic) 0.000000
由上图可知回归方程为:
Y t*=35.97761+0.668695X t*
Se=(8.103546)(0.020642)
t=(4.439737)(32.39512)
R2=0.984983 F=1049.444 DW=1.830746
式中, Yt*=Y t-0.657352Y t-1 , X t*=X t-0.657352X t-1
由于使用了广义差分数据,样本容量减少了 1 个,为 18 个。查 5% 显著水平的 DW 统计表
可知, dL=1.158,d U =1.391 模型中 DW=1,830746 ,du
本文档为【计量经济学庞皓第三版课后答案】,请使用软件OFFICE或WPS软件打开。作品中的文字与图均可以修改和编辑,
图片更改请在作品中右键图片并更换,文字修改请直接点击文字进行修改,也可以新增和删除文档中的内容。
该文档来自用户分享,如有侵权行为请发邮件ishare@vip.sina.com联系网站客服,我们会及时删除。
[版权声明] 本站所有资料为用户分享产生,若发现您的权利被侵害,请联系客服邮件isharekefu@iask.cn,我们尽快处理。
本作品所展示的图片、画像、字体、音乐的版权可能需版权方额外授权,请谨慎使用。
网站提供的党政主题相关内容(国旗、国徽、党徽..)目的在于配合国家政策宣传,仅限个人学习分享使用,禁止用于任何广告和商用目的。