影响GDP增长的经济因素分析
组员：苏敏（分析、撰文）；孙戎（协助撰文、编辑排版、搜集资料）；曾炯、李黎、蒋文（搜集资料）
近年来，我国GDP逐年增长，经济发展速度令人瞩目。为更好的了解我国经济增长的原因，我组对影响我国GDP增长的经济因素进行了分析。
下表（表1.1）提供了我国1978——2002年的GDP及其主要影响因素的数据。其中Y=GDP(亿元)；X1＝能源消费总量（万吨标准煤）；X2＝就业人员（万人）；X3=居民消费水平（元）；X4＝农业总产值（亿元）；X5＝社会消费品零售总额（亿元）；X6＝进出口贸易总额（亿元）
Obs X1 X2 X3 X4 X5 X6 Y
1978 57144 40152 184 1397 1558.6 355 3624.1
1979 58588 41024 197 1697.6 1800 454.6 4038.2
1980 60275 42361 236 1922.6 2140 570 4517.8
1981 59447 43725 249 2180.62 2350 735.3 4862.4
1982 62067 45295 266 2483.26 2570 771.3 5294.7
1983 66040 46436 289 2750 2849.4 860.1 5934.5
1984 70904 48197 327 3214.13 3376.4 1201 7171
1985 76682 49873 437 3619.49 4305 2066.7 8964.4
1986 80850 51282 447 4013.01 4950 2850.4 10202.2
1987 86632 52783 508 4675.7 5820 3084.2 11962.5
1988 92997 54334 635 5865.27 7440 3822 14928.3
1989 96934 55329 762 6534.73 8101.4 4155.9 16909.2
1990 98703 56740 803 7662.09 8300.1 5560.1 18547.9
1991 103783 58360 896 8157.03 9415.6 7229.3 21617.8
1992 109170 59432 1070 9084.7 10993.7 9119.6 26638.1
1993 115993 60220 1331 10995.5 12462.1 11271 34634.4
1994 122737 61470 1746 15750.5 16264.7 20381.9 46759.4
1995 131176 62388 2236 20340.9 20620 23499.9 58478.1
1996 138948 68850 2641 22353.7 24774.1 24133.8 67884.6
1997 138173 69600 2834 23788.4 27298.9 26967.2 74462.6
1998 132214 70637 2972 24541.9 29152.5 26849.7 78345.2
1999 130119 71394 3138 24519.1 31134.7 29896.2 82067.5
2000 130297 72085 3397 32917.93 34152.6 39273.2 89468.1
2001 134914 73025 3609 37213.49 37595.2 42183.6 97314.8
2002 148222 73740 3791 43499.91 40910.5 51378.2 104790.6
一：现估计模型为Y=A0+A1*X1+A2*X2+A3*X3+A4*X4+A5*X5+A6*X6+U 运用OLS估计方法对上式中得参数进行估计，利用Eviews软件得回归分析结果如下：
（表1.2）
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 20:44
Sample: 1978 2002
Included observations: 25
Variable Coefficient Std. Error t-Statistic Prob.
C 5421.215 3244.856 1.670710 0.1121
X1 0.050933 0.030215 1.685708 0.1091
X2 -0.224428 0.113375 -1.979516 0.0633
X3 21.18387 2.084941 10.16042 0.0000
X4 -0.216535 0.203877 -1.062084 0.3022
X5 0.488490 0.311803 1.566665 0.1346
X6 0.336620 0.139208 2.418102 0.0264
R-squared 0.999725 Mean dependent var 35976.74
Adjusted R-squared 0.999634 S.D. dependent var 34444.88
S.E. of regression 658.9930 Akaike info criterion 16.05080
Sum squared resid 7816893. Schwarz criterion 16.39208
Log likelihood -193.6350 F-statistic 10925.17
Durbin-Watson stat 1.748019 Prob(F-statistic) 0.000000
分析回归结果：
从经济意义上讲，就业人口X2的系数为负，可初步认为国民经济在向技术密集型、资本密集型发展。农业总产值的系数为负，不符合实际经济意义。其余解释变量的系数为正，符合实际经济现象。
从模型检验上讲，拟合较好。可决系数R^（2）=0.999725,F统计量为10925.17>2.66=F0.05(6,18)表明模型在整体上拟合非常好；系数显著性检验：对于A0,t统计量为1.670710，给定ａ＝0.05 查t分布表，在自由度为n-k=18下，得临界值T0.025(18)=2.101
因为t<T0.025(18),所以接受H0:A0=0的原假设。
对于A1、A2、A3、A4、A5、A6,除去A3、A6的t统计量大于T0.025(18)之外，其余系数的t统计量均小于T0.025(18) ，因此可初步认为模型存在严重的多重共线性。
现重新估计模型为Y=A1X1+A2X2+A3X3+A4X4+A5X5+A6X6
得回归结果如下（表1.3）：
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 20:57
Sample: 1978 2002
Included observations: 25
Variable Coefficient Std. Error t-Statistic Prob.
X1 0.010656 0.019053 0.559269 0.5825
X2 -0.041248 0.030184 -1.366564 0.1877
X3 22.68974 1.966670 11.53714 0.0000
X4 -0.140006 0.207819 -0.673692 0.5086
X5 0.146023 0.245779 0.594124 0.5594
X6 0.387108 0.142150 2.723241 0.0135
R-squared 0.999683 Mean dependent var 35976.74
Adjusted R-squared 0.999599 S.D. dependent var 34444.88
S.E. of regression 689.3576 Akaike info criterion 16.11496
Sum squared resid 9029064. Schwarz criterion 16.40749
Log likelihood -195.4370 Durbin-Watson stat 1.704338
从模型检验上看，R^(2)=0.999683小于第一次模型的可决系数;T检验也并不优于第一次模型的t检验，故仍采用第一次模型。
二、多重共线性检验
1、检验：
利用Eviews计算线性回归模型中，六个解释变量的如下简单相关系数矩阵(表2.1.1)：
X1 X2 X3 X4 X5 X6
X1 1 0.978454327431 0.92085407884 0.888167174119 0.911118877104 0.883472715462
X2 0.978454327431 1 0.948815761328 0.917125867946 0.946036086594 0.909232951166
X3 0.92085407884 0.948815761328 1 0.982786453939 0.997008287295 0.982361794167
X4 0.888167174119 0.917125867946 0.982786453939 1 0.990709311732 0.997396156937
X5 0.911118877104 0.946036086594 0.997008287295 0.990709311732 1 0.98844222336
X6 0.883472715462 0.909232951166 0.982361794167 0.997396156937 0.98844222336 1
从上表可以看出，各解释变量之间存在高度线性相关。同时由表1.2又可看出，尽管整体上线性回归拟合较好，但X1 X2 X4 X5 变量的参数T值并不显著，表明模型中解释变量确实存在严重的多重共线性。
2、修正：
⑴运用OLS方法逐一求出Y对各个解释变量的回归，分别如下：
Y=-67070.34+1.029232X1 （式2.1.1）
(9781.140) (0.093575)
t=(-6.856618) (10.99902)
R^(2)=0.840254 Se=14063.12 F=120.9784
Y=-133299.7+2.962005X2 （式2.1.2）
(12588.50) (0.212476)
t=(-10.58901) (13.68286)
R^(2)=0.890591 Se=11638.38 F=187.2206
Y=-2268.943+27.31756X3 （式2.1.3）
(348.7497) (0.186822)
t=(-6.505936) (146.2225)
R^(2)=0.998925 Se=1153.406 F=21381.06
Y=617.7713+2.752282X4 （式2.1.4）
(1669.79)(0.094620)
t=(0.369969) (29.08787)
R^(2)=0.973536 Se=5723.931 F=846.1039
Y=-1873.193+2.700977X5 （式2.1.5）
(712.2024) (0.037971)
t=(-2.630142) (71.13173)
R^(2)=0.995475 Se=2366.912 F=5059.723
Y=5875.266+2.222034X6 （式2.1.6）
(1531.230) (0.075790)
t=（3.836958）（29.31845）
R^(2)=0.973940 Se=5680.092 F=859.5713
综合分析可见，在七个一元回归模型中，GDP(Y)对居民消费水平(X3)线性关系强，拟合程度好。
（2）将其余解释变量逐一带入对X3的一元线性回归方程中，得以下几个模型：
Y=-387.7386-0.027368X1+27.93105X3 （式2.2.1）
(1367.242) (0.019261) (0.468871)
t=(-0.283592) (-1.420886) (59.57089)
￣R^(2)=0.998926 Se=1128.676 F=11165.14
Y=6262.980+0.029858X1-0.232206X2+28.56686X3 （式2.2.2）
(3643.674) (0.034485) (0.119009) (0.548771)
t=(1.718864) (0.865831) (-1.951160) (52.05602)
￣R^(2)=0.999048 Se=1062.902 F=8394.415
Y=4027.197+0.032089X1-0.181952X2+24.94421X3+0.327880X4 （式2.2.3）
(2612.367) (0.024319) (0.084582) (0.860040) (0.069519)
t=(1.541590) (1.319521) (-2.151190) (29.00354) (4.716419)
￣R^(2)=0.999606 Se=749.4065 F=12670.52
Y=7124.543+0.061735X1-0.295985X2+22.25771X3+0.173648X4+0.442008X5 （式2.2.4）
(3548.587)(0.033478) (0.122613) (2.282200) (0.13910) (0.348654)
t=(2.007713)(1.844017) (-2.413968) (9.753743) (1.243811) (1.267755)
￣R^(2)=0.999541 Se=738.2831 F=10444.48
Y=5421.215+0.050933X1-0.224428X2+21.18387X3-0.216535X4+0.488490X5+0.336620X6 (式2.2.5)
（3244.856）（0.030215）（0.113375）（2.084941）（0.203877）（0.311803）（0.139208）
t=（1.670710）（1.685708）（-1.979516）（10.16042）（-1.062084）（1.566665）（2.418102）
￣R^(2)=0.999634 Se=658.9930 F=10925.17
从式（2.2.1）可以看出，解释变量X1与X2之间存在共线性。又因为X2对Y的经济意义影响低于X1，故舍去X2。
从式（2.2.4）（2.2.5）看出，解释变量X4 X5对Y的影响并不显著，故将X4 X5撤去得如下模型（表2.2.1）：
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 22:06
Sample: 1978 2002
Included observations: 25
Variable Coefficient Std. Error t-Statistic Prob.
C -903.4074 829.1531 -1.089554 0.2883
X1 -0.005261 0.012145 -0.433206 0.6693
X3 23.63361 0.740702 31.90706 0.0000
X6 0.318814 0.050783 6.277944 0.0000
R-squared 0.999658 Mean dependent var 35976.74
Adjusted R-squared 0.999609 S.D. dependent var 34444.88
S.E. of regression 681.1097 Akaike info criterion 16.03097
Sum squared resid 9742120. Schwarz criterion 16.22599
Log likelihood -196.3871 F-statistic 20452.98
Durbin-Watson stat 1.631881 Prob(F-statistic) 0.000000
表2.2.1中能源消费总量x1的系数为负,不符合实际经济意义,现舍去1978至1982年的数据,重新回归如下:
(表2.2.2)
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 22:10
Sample: 1983 2002
Included observations: 20
Variable Coefficient Std. Error t-Statistic Prob.
C -1797.334 1438.251 -1.249666 0.2294
X1 0.004777 0.018343 0.260447 0.7978
X3 23.50009 0.841880 27.91382 0.0000
X6 0.317367 0.056410 5.626057 0.0000
R-squared 0.999589 Mean dependent var 43854.06
Adjusted R-squared 0.999512 S.D. dependent var 34234.35
S.E. of regression 756.2503 Akaike info criterion 16.27148
Sum squared resid 9150632. Schwarz criterion 16.47062
Log likelihood -158.7148 F-statistic 12973.20
Durbin-Watson stat 1.689291 Prob(F-statistic) 0.000000
经过上述逐步回归分析，表明y对x1 x3 x6的回归模型最优。
三、异方差性检验
1、检验
（1）、Goldfeld—Quandt检验
用OLS方法求得下列结果：
Y=-12754.36+0.236669X1+10.96853X3-0.395909X6 （1983——1989）
(6250.887) (0.1134474) (3.689418) (0.767691)
￣R^(2)=0.994022 ∑e1^(2)=287719.1
Y=-5129.215+0.035047X1+23.39430X3+0.304576X6 (1994——2002)
（7786.425）（0.065303）（1.268709）(0.079944)
￣R^(2)=0.996966 ∑e2^(2)=5139285
求F统计量： F=∑e2^(2)/∑e1^(2)=17.86216139
给定显著性水平ａ=0.05，得临界值F0.05(4,4)=4.28,比较F=17.86216139＞ F0.05(4,4)=6.39
可初步认为不存在异方差。
（2）、Arch检验
利用eviews软件输出结果为（表3.1.1）
Dependent Variable: E2
Method: Least Squares
Date: 06/03/05 Time: 23:01
Sample(adjusted): 1986 2002
Included observations: 17 after adjusting endpoints
Variable Coefficient Std. Error t-Statistic Prob.
C 602213.2 337049.7 1.786719 0.0973
E2(-1) -0.299677 0.304742 -0.983380 0.3434
E2(-2) 0.087203 0.289528 0.301188 0.7680
E2(-3) -0.005035 0.542470 -0.009282 0.9927
R-squared 0.115727 Mean dependent var 495758.6
Adjusted R-squared -0.088335 S.D. dependent var 699258.1
S.E. of regression 729489.3 Akaike info criterion 30.04040
Sum squared resid 6.92E+12 Schwarz criterion 30.23645
Log likelihood -251.3434 F-statistic 0.567116
Durbin-Watson stat 1.948353 Prob(F-statistic) 0.646348
丛输出的辅助回归函数重的R^(2)计算（n-p）*R^(2)=17*0.115727=1.967359,查χ^(2)分布表
给定α=0.05得临界值χ^(2)0.05（3）=7.81因为（n-p）*R^(2)=17*0.115727=1.967359＜χ^(2)0.05（3）=7.81所以拒绝H0,表明不存在异方差。
四、自相关检验
1、检验
（1）、图示法
由表2.2.2的OLS估计可直接得到残差resid,生成序列E,输出结果如下图：
由此图可以看出，残差et不成线性自回归，可初步认为随机误差ut不存在自相关。
（2）DW检验
根据表2.2.2估计的结果，DW=1.689291.给定显著性水平α=0.05，查DW表n=20,k=3得下限临界值dl=0.998
上限临界值du=1.676
因为du<DW统计量=1.689291<4-du,表明不存在自相关。
五、运用阿尔蒙法进行滞后性修正
1，对Y ,X1估计如下有限分布滞后模型
Y=A0+B0X1+B1X1（-1）+B2X1(-2)+B3X1(-3)
应用EVIEWS软件得回归分析结果如下：
Dependent Variable: Y
Method: Least Squares
Date: 06/04/05 Time: 23:27
Sample: 1983 2002
Included observations: 20
Variable Coefficient Std. Error t-Statistic Prob.
C -78223.17 10398.98 -7.522195 0.0000
PDL01 -0.914296 0.400006 -2.285705 0.0362
PDL02 -0.828224 0.410601 -2.017099 0.0608
PDL03 1.087286 0.388824 2.796349 0.0129
R-squared 0.935158 Mean dependent var 43854.06
Adjusted R-squared 0.923001 S.D. dependent var 34234.35
S.E. of regression 9499.609 Akaike info criterion 21.33275
Sum squared resid 1.44E+09 Schwarz criterion 21.53189
Log likelihood -209.3275 F-statistic 76.91841
Durbin-Watson stat 0.481394 Prob(F-statistic) 0.000000
Lag Distribution of X1 i Coefficient Std. Error T-Statistic
. * | 0 1.00121 0.44552 2.24728
* . | 1 -0.91430 0.40001 -2.28570
* . | 2 -0.65523 0.39089 -1.67625
. *| 3 1.77840 0.45803 3.88274
Sum of Lags 1.21009 0.08629 14.0238
即是Y＝-78223.17+1.00121X1－0.91430X1(-1) -0.65523X1(-2)+ 1.77840X1(-3)
2对Y ,X3有限分布滞后模型
Y=A0+B0X3+B1X3(-1)+B2X3(-2)+B3X3(-3)
用EVIEWS软件显示回归分如下：
Dependent Variable: Y
Method: Least Squares
Date: 06/04/05 Time: 23:31
Sample: 1983 2002
Included observations: 20
Variable Coefficient Std. Error t-Statistic Prob.
C -2497.770 516.2504 -4.838292 0.0002
PDL01 2.106917 2.176725 0.967930 0.3475
PDL02 -15.95829 2.156581 -7.399811 0.0000
PDL03 8.550866 2.150222 3.976737 0.0011
R-squared 0.998874 Mean dependent var 43854.06
Adjusted R-squared 0.998663 S.D. dependent var 34234.35
S.E. of regression 1251.806 Akaike info criterion 17.27942
Sum squared resid 25072277 Schwarz criterion 17.47856
Log likelihood -168.7942 F-statistic 4731.442
Durbin-Watson stat 1.165284 Prob(F-statistic) 0.000000
Lag Distribution of X3 i Coefficient Std. Error T-Statistic
. *| 0 26.6161 2.22221 11.9773
.* | 1 2.10692 2.17672 0.96793
* . | 2 -5.30051 2.11824 -2.50232
. * | 3 4.39380 2.45928 1.78662
Sum of Lags 27.8163 0.33326 83.4682
即是Y＝-2497.770+26.6161 X3+2.10692 X3 (-1) -5.30051 X3 (-2)+ 4.39380 X3 (-3)
3对Y , x6有限分布滞后模型
Y=A0+B0 x6+B1 x6 (-1)+B2 x6 (-2)+B3 x6 (-3)
用EVIEWS软件显示回归分如下：
Dependent Variable: Y
Method: Least Squares
Date: 06/04/05 Time: 23:37
Sample: 1983 2002
Included observations: 20
Variable Coefficient Std. Error t-Statistic Prob.
C 8207.606 1631.510 5.030680 0.0001
PDL01 0.774028 0.343247 2.255015 0.0385
PDL02 -0.075852 0.363485 -0.208680 0.8373
PDL03 -0.064975 0.330846 -0.196391 0.8468
R-squared 0.983441 Mean dependent var 43854.06
Adjusted R-squared 0.980336 S.D. dependent var 34234.35
S.E. of regression 4800.586 Akaike info criterion 19.96772
Sum squared resid 3.69E+08 Schwarz criterion 20.16687
Log likelihood -195.6772 F-statistic 316.7499
Durbin-Watson stat 0.342226 Prob(F-statistic) 0.000000
Lag Distribution of X6 i Coefficient Std. Error T-Statistic
. *| 0 0.78491 0.39271 1.99867
. *| 1 0.77403 0.34325 2.25501
. * | 2 0.63320 0.34422 1.83951
. * | 3 0.36242 0.46095 0.78625
Sum of Lags 2.55456 0.14135 18.0730
即是Y=8207.606+0.78491 x6+0.77403 x6 (-1)+ 0.63320 x6 (-2)+ 0.36242 x6 (-3)
通过以上一系列统计检验可以说明：我国GDP的增长与能源消费总量X1,居民消费水平X3，进出口贸易总额X6有很高的相关性。其中，又以居民消费水平X3的影响程度最为显著。由此可以看出影响我国GDP的主要因素是居民消费水平，进出口贸易总额，能源消费总量。
附本（相关表格）
（表2.1.2）
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 21:08
Sample: 1978 2002
Included observations: 25
Variable Coefficient Std. Error t-Statistic Prob.
C -67070.34 9781.840 -6.856618 0.0000
X1 1.029232 0.093575 10.99902 0.0000
R-squared 0.840254 Mean dependent var 35976.74
Adjusted R-squared 0.833308 S.D. dependent var 34444.88
S.E. of regression 14063.12 Akaike info criterion 22.01712
Sum squared resid 4.55E+09 Schwarz criterion 22.11463
Log likelihood -273.2140 F-statistic 120.9784
Durbin-Watson stat 0.094594 Prob(F-statistic) 0.000000
(表2.1.3)
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 21:12
Sample: 1978 2002
Included observations: 25
Variable Coefficient Std. Error t-Statistic Prob.
C -133299.7 12588.50 -10.58901 0.0000
X2 2.962005 0.216476 13.68286 0.0000
R-squared 0.890591 Mean dependent var 35976.74
Adjusted R-squared 0.885834 S.D. dependent var 34444.88
S.E. of regression 11638.38 Akaike info criterion 21.63862
Sum squared resid 3.12E+09 Schwarz criterion 21.73613
Log likelihood -268.4828 F-statistic 187.2206
Durbin-Watson stat 0.159690 Prob(F-statistic) 0.000000
(表2。1.4)
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 21:15
Sample: 1978 2002
Included observations: 25
Variable Coefficient Std. Error t-Statistic Prob.
C -2268.943 348.7497 -6.505936 0.0000
X3 27.31756 0.186822 146.2226 0.0000
R-squared 0.998925 Mean dependent var 35976.74
Adjusted R-squared 0.998879 S.D. dependent var 34444.88
S.E. of regression 1153.406 Akaike info criterion 17.01544
Sum squared resid 30597948 Schwarz criterion 17.11295
Log likelihood -210.6931 F-statistic 21381.06
Durbin-Watson stat 0.841840 Prob(F-statistic) 0.000000
(表2.1.5)
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 21:17
Sample: 1978 2002
Included observations: 25
Variable Coefficient Std. Error t-Statistic Prob.
C 617.7713 1669.790 0.369969 0.7148
X4 2.752282 0.094620 29.08787 0.0000
R-squared 0.973536 Mean dependent var 35976.74
Adjusted R-squared 0.972385 S.D. dependent var 34444.88
S.E. of regression 5723.931 Akaike info criterion 20.21932
Sum squared resid 7.54E+08 Schwarz criterion 20.31683
Log likelihood -250.7415 F-statistic 846.1039
Durbin-Watson stat 0.557749 Prob(F-statistic) 0.000000
( 表2.1.6)
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 21:21
Sample: 1978 2002
Included observations: 25
Variable Coefficient Std. Error t-Statistic Prob.
C -1873.193 712.2024 -2.630142 0.0150
X5 2.700977 0.037971 71.13173 0.0000
R-squared 0.995475 Mean dependent var 35976.74
Adjusted R-squared 0.995278 S.D. dependent var 34444.88
S.E. of regression 2366.912 Akaike info criterion 18.45318
Sum squared resid 1.29E+08 Schwarz criterion 18.55069
Log likelihood -228.6647 F-statistic 5059.723
Durbin-Watson stat 0.288053 Prob(F-statistic) 0.000000
(表2.1.7)
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 21:23
Sample: 1978 2002
Included observations: 25
Variable Coefficient Std. Error t-Statistic Prob.
C 5875.266 1531.230 3.836958 0.0008
X6 2.222034 0.075790 29.31845 0.0000
R-squared 0.973940 Mean dependent var 35976.74
Adjusted R-squared 0.972807 S.D. dependent var 34444.88
S.E. of regression 5680.092 Akaike info criterion 20.20394
Sum squared resid 7.42E+08 Schwarz criterion 20.30145
Log likelihood -250.5493 F-statistic 859.5713
Durbin-Watson stat 0.739809 Prob(F-statistic) 0.000000
（表2.2.1）
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 21:28
Sample: 1978 2002
Included observations: 25
Variable Coefficient Std. Error t-Statistic Prob.
C -387.7386 1367.242 -0.283592 0.7794
X1 -0.027368 0.019261 -1.420886 0.1694
X3 27.93105 0.468871 59.57089 0.0000
R-squared 0.999016 Mean dependent var 35976.74
Adjusted R-squared 0.998926 S.D. dependent var 34444.88
S.E. of regression 1128.676 Akaike info criterion 17.00765
Sum squared resid 28026027 Schwarz criterion 17.15391
Log likelihood -209.5956 F-statistic 11165.14
Durbin-Watson stat 0.973989 Prob(F-statistic) 0.000000
(表2.2.2)
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 21:31
Sample: 1978 2002
Included observations: 25
Variable Coefficient Std. Error t-Statistic Prob.
C 6262.980 3643.674 1.718864 0.1003
X1 0.029858 0.034485 0.865831 0.3964
X2 -0.232206 0.119009 -1.951160 0.0645
X3 28.56686 0.548771 52.05602 0.0000
R-squared 0.999167 Mean dependent var 35976.74
Adjusted R-squared 0.999048 S.D. dependent var 34444.88
S.E. of regression 1062.902 Akaike info criterion 16.92104
Sum squared resid 23724995 Schwarz criterion 17.11606
Log likelihood -207.5130 F-statistic 8394.415
Durbin-Watson stat 0.984561 Prob(F-statistic) 0.000000
(表2.2.3)
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 21:33
Sample: 1978 2002
Included observations: 25
Variable Coefficient Std. Error t-Statistic Prob.
C 4027.197 2612.367 1.541590 0.1388
X1 0.032089 0.024319 1.319521 0.2019
X2 -0.181952 0.084582 -2.151190 0.0439
X3 24.94421 0.860040 29.00354 0.0000
X4 0.327880 0.069519 4.716419 0.0001
R-squared 0.999606 Mean dependent var 35976.74
Adjusted R-squared 0.999527 S.D. dependent var 34444.88
S.E. of regression 749.4065 Akaike info criterion 16.25330
Sum squared resid 11232201 Schwarz criterion 16.49707
Log likelihood -198.1662 F-statistic 12670.52
Durbin-Watson stat 1.562943 Prob(F-statistic) 0.000000
(表2.2.4)
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 21:40
Sample: 1978 2002
Included observations: 25
Variable Coefficient Std. Error t-Statistic Prob.
C 7124.543 3548.587 2.007713 0.0591
X1 0.061735 0.033478 1.844017 0.0808
X2 -0.295985 0.122613 -2.413968 0.0260
X3 22.25771 2.282200 9.752743 0.0000
X4 0.173648 0.139610 1.243811 0.2287
X5 0.442008 0.348654 1.267755 0.2202
R-squared 0.999636 Mean dependent var 35976.74
Adjusted R-squared 0.999541 S.D. dependent var 34444.88
S.E. of regression 738.2831 Akaike info criterion 16.25209
Sum squared resid 10356175 Schwarz criterion 16.54463
Log likelihood -197.1512 F-statistic 10444.48
Durbin-Watson stat 1.408484 Prob(F-statistic) 0.000000
(表3.1.1)
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 22:21
Sample: 1983 1989
Included observations: 7
Variable Coefficient Std. Error t-Statistic Prob.
C -12754.36 6250.887 -2.040409 0.1340
X1 0.236669 0.113474 2.085666 0.1283
X3 10.96853 3.689418 2.972970 0.0589
X6 -0.395909 0.767691 -0.515715 0.6417
R-squared 0.997011 Mean dependent var 10867.44
Adjusted R-squared 0.994022 S.D. dependent var 4005.294
S.E. of regression 309.6875 Akaike info criterion 14.60456
Sum squared resid 287719.1 Schwarz criterion 14.57366
Log likelihood -47.11597 F-statistic 333.5425
Durbin-Watson stat 2.231426 Prob(F-statistic) 0.000277
(表3.1.2)
Dependent Variable: Y
Method: Least Squares
Date: 06/03/05 Time: 22:31
Sample: 1994 2002
Included observations: 9
Variable Coefficient Std. Error t-Statistic Prob.
C -5129.215 7786.425 -0.658738 0.5392
X1 0.035047 0.065303 0.536687 0.6145
X3 23.39430 1.268709 18.43945 0.0000
X6 0.304576 0.079944 3.809862 0.0125
R-squared 0.998104 Mean dependent var 77730.10
Adjusted R-squared 0.996966 S.D. dependent var 18405.97
S.E. of regression 1013.833 Akaike info criterion 16.98197
Sum squared resid 5139285. Schwarz criterion 17.06962
Log likelihood -72.41885 F-statistic 877.2615
Durbin-Watson stat 2.005411 Prob(F-statistic) 0.000000
（表4.2.1）
Dependent Variable: DY
Method: Least Squares
Date: 05/07/05 Time: 16:55
Sample(adjusted): 1979 2002
Included observations: 24 after adjusting endpoints
Variable Coefficient Std. Error t-Statistic Prob.
DX2 -0.032787 0.008348 -3.927657 0.0008
DX3 25.19594 1.014343 24.83966 0.0000
DX6 0.177588 0.089104 1.993033 0.0594
R-squared 0.998873 Mean dependent var -44710.72
Adjusted R-squared 0.998765 S.D. dependent var 44968.62
S.E. of regression 1580.162 Akaike info criterion 17.68491
Sum squared resid 52435167 Schwarz criterion 17.83217
Log likelihood -209.2189 Durbin-Watson stat 1.926203