中国粮食产量的影响因素研究—计量经济学论文

来源：用户分享时间：2025/7/26 17:19:08 本文由

说明：文章内容仅供预览，部分内容可能不全，需要完整文档或者需要复制内容，请下载word后使用。下载word有问题请添加微信号:xxxxxxx或QQ：xxxxxx 处理（尽可能给您提供完整文档），感谢您的支持与谅解。

当设置Period的Effects Specification 为None，Cross-section为Random时做Correlated Random Effects-Hausam tests 可得P=0<0.5 所以就拒绝Cross-section random

具体检验过程如下所示：

Correlated Random Effects - Hausman Test Equation: Untitled

Test cross-section random effects

Test Summary Cross-section random

Variable SUN TEMP RAIN SEED LABOR

Dependent Variable: OUTPUT Method: Panel Least Squares Date: 01/02/12 Time: 18:22 Sample: 2003 2009 Periods included: 7 Cross-sections included: 31

Total panel (balanced) observations: 217

Variable C SUN TEMP RAIN SEED LABOR

Coefficient 1493.784 0.085407 2.580547 0.010962 0.439032 -1.660357

Std. Error 277.5476 0.033183 13.74200 0.032484 0.026036 0.121743

t-Statistic 5.382083 2.573832 0.187785 0.337450 16.86263 -13.63817

Fixed 0.085407 2.580547 0.010962 0.439032 -1.660357

Chi-Sq. Statistic 123.137431

Random 0.039828 38.310651 0.022599 0.551235 -0.636000

Chi-Sq. d.f.

Var(Diff.) 0.000052 104.135597 0.000044 0.000327 0.009351

Prob. 0.0000 Prob. 0.0000 0.0005 0.0795 0.0000 0.0000 Prob. 0.0000 0.0109 0.8513 0.7362 0.0000 0.0000

Cross-section random effects test comparisons:

Cross-section random effects test equation:

Effects Specification

Cross-section fixed (dummy variables) R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic)

0.994413 Mean dependent var 0.993333 S.D. dependent var 101.7104 Akaike info criterion 1872447. Schwarz criterion -1291.230 Hannan-Quinn criter. 920.4879 Durbin-Watson stat 0.000000

1586.542 1245.656 12.23253 12.79325 12.45904 1.902462

因而可以假定Period的Effects Specification 为Fix ，Cross-section为Fix，从而做回归。

②通过eviews的hausman检验，可得出Cross-section与Period都拒绝random，则可假定Cross-section与Period都为fixed。若这样假定，可得到回归的结果如下：

c(1) c(2) c(3) c(4) c(5) c(6) 系数 0.71 9.89 0.44 -1.66 T值 2.08 0.63 16.3 -11.21 1433.706 4.86 -0.00048 -0.01 并且可得：R^2=0.99 F=798.66 由此可得Output=1433.706+0.71*sun+9.89*temp-0.00048*rain+0.44*seed -1.66*labor

根据此次的回归结果，首先可以知道各个系数及其符号所表示的情况，（1）1433.706说明当其他影响因素都为0 时，粮食产量为1433.706万吨，这表示当这几个解释变量都为0时依然有粮食产量1433.706万吨，这种情况可能是由于自然界中一些野生顽强的粮食作物品种导致的；（2）光照小时数、气温、播种面积都对粮食产量具有正面的影响效果，其中以播种面积对于粮食产量的正面影响最大；（3）降雨量以及劳动力对粮食产量存在负面影响。

其次，我们可以看到（1）R-squared=0.99，说明此模型的拟合程度较好，各个解释变量可以较好地解释被解释变量；（2）F-statistic=798.66，说明整个方程是显著的。（3）c(1)、c(2)、c(5)、c(6)各系数对应的t值可知，sun、seed和labor这几个变量对于粮食产量output的影响都是显著的；而由c(3)、c(4)对应的t值可知，temp和rain对于粮食产量output的影响是不显著的。

具体水平回归模型如下：

Dependent Variable: OUTPUT Method: Panel Least Squares Date: 01/02/12 Time: 18:25 Sample: 2003 2009 Periods included: 7 Cross-sections included: 31

Total panel (balanced) observations: 217

Variable C SUN TEMP RAIN SEED LABOR

Coefficient 1433.706 0.071013 9.892028 -0.000481 0.437261 -1.661748

Std. Error 294.7470 0.034100 15.62226 0.033520 0.026823 0.148260

t-Statistic 4.864192 2.082508 0.633201 -0.014354 16.30148 -11.20833

Prob. 0.0000 0.0388 0.5274 0.9886 0.0000 0.0000 1586.542 1245.656 12.23815 12.89232 12.50240 1.817543

Effects Specification

Cross-section fixed (dummy variables) Period fixed (dummy variables) R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic)

0.994684 Mean dependent var 0.993439 S.D. dependent var 100.9012 Akaike info criterion 1781682. Schwarz criterion -1285.839 Hannan-Quinn criter. 798.6550 Durbin-Watson stat 0.000000

2、对于上述的各变量，建立如下针对参数而言的对数回归模型：

Output=c(1)+c(2)*log（sun）+c(3)*temp +c(4)*rain +c(5)*log（seed）+c(6)*labor +u①hausman检验

当设置Period的Effects Specification 为Random ，Cross-section为None时做Correlated Random Effects-Hausam tests 可得P=0.049<0.5 所以就拒绝Period random 具体检验如下：

Correlated Random Effects - Hausman Test Equation: EQ04

Test period random effects

Test Summary Period random

Chi-Sq. Statistic 11.112193

Chi-Sq. d.f.

Prob. 0.0492

Random

Var(Diff.)

Prob.

** WARNING: estimated period random effects variance is zero.

Variable

Fixed

Period random effects test comparisons:

LOG(SUN) TEMP RAIN LOG(SEED) LABOR

394.617271 -58.365088 0.308959 441.662414 0.965608

411.005605 -55.239832 0.303206 460.079764 0.930991

280.812175 4.602988 0.000426 35.320171 0.000121

0.3281 0.1452 0.7805 0.0019 0.0017 Prob. 0.0004 0.0188 0.0000 0.0139 0.0000 0.0000 1586.542 1245.656 15.56569 15.75260 15.64119 0.107467

Period random effects test equation: Dependent Variable: OUTPUT Method: Panel Least Squares Date: 01/02/12 Time: 18:33 Sample: 2003 2009 Periods included: 7 Cross-sections included: 31

Total panel (balanced) observations: 217

Variable C LOG(SUN) TEMP RAIN LOG(SEED) LABOR

Coefficient -5121.095 394.6173 -58.36509 0.308959 441.6624 0.965608

Std. Error 1424.638 166.5911 12.73754 0.124497 52.61723 0.091706

t-Statistic -3.594665 2.368778 -4.582133 2.481665 8.393874 10.52937

Effects Specification

Period fixed (dummy variables) R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic)

0.804658 Mean dependent var 0.794176 S.D. dependent var 565.1270 Akaike info criterion 65470540 Schwarz criterion -1676.877 Hannan-Quinn criter. 76.76738 Durbin-Watson stat 0.000000

当设置Period的Effects Specification 为None，Cross-section为Random时做Correlated Random Effects-Hausam tests 可得P=0<0.5 所以就拒绝Cross-section random 具体检验如下：

Correlated Random Effects - Hausman Test Equation: EQ04

Test cross-section random effects

搜索更多关于：中国粮食产量的影响因素研究—计量经济学论文的文档

中国粮食产量的影响因素研究—计量经济学论文.doc 将本文的Word文档下载到电脑，方便复制、编辑、收藏和打印

下载这篇word文档

本文链接：https://www.diyifanwen.net/c6qgb75ow3s85bn78adf4_3.html（转载请注明文章来源）