我国农民收入影响因素的回归分析

1、我国农民收入影响因素的回归分析自改革开放以来 , 虽然中国经济平均增长速度为 9.5 % , 但二元经济结构给 经济发展带来的问题仍然很突出。农村人口占了中国总人口的 70 %多, 农业产业 结构不合理 ,经济不发达 ,以及农民收入增长缓慢等问题势必成为我国经济持续 稳定增长的障碍。正确有效地解决好“三农”问题是中国经济走出困境, 实现长期稳定增长的关键。其中 ,农民收入增长是核心 , 也是解决“三农”问题的关键。 本文力图应用适当的多元线性回归模型 , 对有关农民收入的历史数据和现状进行 分析,寻找其根源 ,探讨影响农民收入的主要因素 ,并在此基础上对如何增加农民 收入提出相应的政策建议。农

2、民收入水平的度量， 通常采用人均纯收入指标。 影响农民收入增长的因 素是多方面的， 既有结构性矛盾因素， 又有体制性障碍因素。 但可以归纳为以下 几个方面： 一是农产品收购价格水平。 目前农业收入仍是中西部地区农民收入的 主要来源。 二是农业剩余劳动力转移水平。 中国的农业目前仍以农户分散经营为 主，农业比较效益低， 尽快地把农业剩余劳动力转移出去是有效改善农民收入状 况的重要因素。三是城市化、工业化水平。中国多数地区城市化、工业化水平落 后于世界平均水平， 这种状况极大地影响了农民收入的增长。 四是农业产业结构 状况。农林牧渔业对农民收入增长贡献率是不同的。随着我国“入世”后农产品 市场的开

3、放和人民生活水平的提高、 农产品需求市场的改变， 农业结构状况直接 影响着农民收入的增长。 五是农业投入水平。 农民收入与财政农业支出、 农村集 体投入、农户个人投入以及信贷投入都有显著的正相关关系。 农业投入是农民收 入增长的重要保证。 但考虑到农业投入主体的多元性， 既有国家、 集体和农户的 投入，又有银行、企业和外资的投入，考虑到复杂性和可行性，所以对农业投入 与农民收入，本文暂不作讨论。因此，以全国为例，把农民收入与各影响因素关 系进行线性回归分析，并建立数学模型。一、计量经济模型分析（ 一 ） 、数据搜集根据以上分析，我们在影响农民收入因素中引入 7个解释变量。即： x2 -财政用于

4、农业的支出的比重，X3 -第二、三产业从业人数占全社会从业人数的比重,X4 -非农村人口比重，X5 -乡村从业人员占农村人口的比重，X6 -农业总产值占农林牧总产值的比重，X7 -农作物播种面积，X8 -农村用电量。yx2x3x4x5x6x7x8年份:78年可比价比重%比重比重千公顷亿千瓦时1986133.6013.4329.5017.9236.0179.99150104.07253.101987137.6312.2031.3019.3938.6275.63146379.53320.801988 J147.867.6637.60 :23.7145.90P 69.25143625.87508.9

5、01989196.769.4239.9026.2149.2362.75146553.93790.501990220.539.9839.9026.4149.9364.66148362.27844.501991223.2510.2640.30 :26.9450.92:63.09149585.80963.201992 J233.1910.0541.50 :27.4651.53：61.51149007.101106.901993265.679.4943.6027.9951.8660.07147740.701244.901994 335.169.2045.7028.5152.12:58.22148240

6、.601473.901995411.298.4347.80：29.0452.41:58.43149879.301655.701996460.688.8249.5030.4853.2360.57152380.601812.701997477.968.3050.1031.9154.9358.23153969.201980.101998474.0210.6950.20 :33.3555.84:58.03155705.702042.201999466.808.2349.9034.7857.1657.53156372.812173.452000466.167.7550.0036.2259.3355.68

7、156299.852421.302001；469.80：7.7150.00 :37.6660.62:55.24155707.862610.782002468.957.1750.0039.0962.0254.51154635.512993.402003476.24p.1250.9040.5363.7250.08152414.963432.922004499.399.6753.10 :41.7665.64:50.05153552.553933.032005 1521.207.2255.2042.9967.5949.72155487.734375.70资料来源中国统计年鉴 2006。（二）、计量经济

8、学模型建立我们设定模型为下面所示的形式：Y = 12X23X34X45X56X67X7：8乂8 Ut利用Eviews软件进行最小二乘估计，估计结果如下表所示：Depe ndent Variable: YMethod: Least SquaresDate: 06/08/07 Time: 21:51Sample: 1986 2004In cluded observati ons: 19VariableCoefficie ntStd. Errort-StatisticProb.-1102.373375.8283-2.9331840.0136X2-6.6353933.781349-1.7547690.

9、1071X318.229422.0666178.8208990.0000X42.4300398.3703370.2903160.7770X5-16.237375.894109-2.7548470.0187X6-2.1552082.770834-0.7778190.4531X70.0099620.0023284.2788100.0013X80.0633890.0212762.9793480.0125R-squared0.995823Mean depe ndent var345.5232Adjusted R-squared0.993165S.D.dependent var139.7117S.E.

10、of regressi on11.55028Akaike info criteri on8.026857Sum squared resid1467.498Schwarz criteri on8.424516Log likelihood-68.25514F-statistic374.6600Durbi n- Watson stat1.993270Prob(F-statistic)0.000000表1最小二乘估计结果回归分析报告为:Y = -1102.373-6.6354焉 + 18.2294X3 +2.4300X36.2374X5 -2.1552X6 +0.0100X7 +0.0634X8SE

11、=375.833.78132.066618.370345.89412.77080.002330.02128t 二 -2.933-1.7558.820900.20316-2.755-0.7784.278812.97932 2R =0.995823 R =0.993165 Df =19 DW =1.99327 F =374.66、计量经济学检验(一)、多重共线性的检验及修正、检验多重共线性(a)、直观法从“表1最小二乘估计结果”中可以看出，虽然模型的整体拟合的很好,但是x4 x6的t统计量并不显著，所以可能存在多重共线性。(b)、相关系数矩阵从“表2相关系数矩阵”中可以看出，个个解释变量之间的相关

12、程度较高,X2X3X4X5X6X7X8X21.000000-0.717662-0.695257-0.7313260.737028-0.332435-0.594699X3-0.7176621.0000000.9222860.935992-0.9457010.7422510.883804X4-0.6952570.9222861.0000000.986050-0.9377510.7539280.974675X5-0.7313260.9359920.9860501.000000-0.9747500.6874390.940436X60.737028-0.945701-0.937751-0.9747501

13、.000000-0.603539-0.887428X7-0.3324350.7422510.7539280.687439-0.6035391.0000000.742781X8-0.5946990.8838040.9746750.940436-0.8874280.7427811.000000表2相关系数矩阵所以应该存在多重共线性。、多重共线性的修正一一逐步迭代法A、一元回归Depe ndent Variable: YMethod: Least SquaresDate: 06/08/07 Time: 21:52Sample: 1986 2004In cluded observati ons: 19

14、VariableCoefficie ntStd. Errort-StatisticProb.C820.3133151.87125.4013740.0000X2-51.3783616.18923-3.1736140.0056R-squared0.372041Mean depe ndent var345.5232Adjusted R-squared0.335102S.D.dependent var139.7117S.E. of regressi on113.9227Akaike info criteri on12.40822Sum squared resid220632.4Schwarz crit

15、eri on12.50763Log likelihood-115.8781F-statistic10.07183Durbi n- Watson stat0.644400Prob(F-statistic)0.005554表3 y对x2的回归结果Depe ndent Variable: YMethod: Least SquaresDate: 06/08/07 Time: 21:52Sample: 1986 2004In cluded observati ons: 19VariableCoefficie ntStd. Errort-StatisticProb.C-525.889164.11333-8

16、.2024920.0000X319.460311.41604313.742740.0000R-squared0.917421Mean depe ndent var345.5232Adjusted R-squared0.912563S.D.dependent var139.7117S.E. of regressi on41.31236Akaike info criteri on10.37950Sum squared resid29014.09Schwarz criteri on10.47892Log likelihood-96.60526F-statistic188.8628Durbi n- W

17、atson stat0.598139Prob(F-statistic)0.000000表4 y对x3的回归结果Depe ndent Variable: YMethod: Least SquaresDate: 06/08/07 Time: 21:52Sample: 1986 2004In cluded observati ons: 19VariableCoefficie ntStd. Errort-StatisticProb.C-223.190569.92322-3.1919370.0053X418.650862.2422408.3179560.0000R-squared0.802758Mea

18、n depe ndent var345.5232Adjusted R-squared0.791155S.D.dependent var139.7117S.E. of regressi on63.84760Akaike info eriteri on11.25018Sum squared resid69300.77Sehwarz eriteri on11.34959Log likelihood-104.8767F-statistie69.18839Durbi n- Watson stat0.282182Prob(F-statistie)0.000000表5 y对x4的回归结果Depe ndent

19、 Variable: YMethod: Least SquaresDate: 06/08/07Time: 21:52Sample: 1986 2004In eluded observati ons: 19VariableCoeffieie ntStd. Errort-StatistieProb.C-494.1440118.1449-4.1825260.0006X515.779782.1987117.1768320.0000R-squared0.751850Mean depe ndent var345.5232Adjusted R-squared0.737253S.D.dependent var

20、139.7117S.E. of regressi on71.61463Akaike info eriteri on11.47978Sum squared resid87187.14Sehwarz eriteri on11.57919Log likelihood-107.0579F-statistie51.50691Durbi n- Watson stat0.318959Prob(F-statistie)0.000002表6 y对x5的回归结果Depe ndent Variable: YMethod: Least SquaresDate: 06/08/07Time: 21:52Sample: 1

21、986 2004In eluded observati ons: 19VariableCoeffieie ntStd. Errort-StatistieProb.C1288.009143.80888.9563950.0000X6-15.523982.351180-6.6026350.0000R-squared0.719448Mean depe ndent var345.5232Adjusted R-squared0.702945S.D.dependent var139.7117S.E. of regressi on76.14674Akaike info eriteri on11.60250Su

22、m squared resid98571.54Sehwarz eriteri on11.70192Log likelihood-108.2238F-statistie43.59479Durbi n- Watson stat0.395893Prob(F-statistie)0.000004表7 y对x6的回归结果Depe ndent Variable: YMethod: Least SquaresDate: 06/08/07Time: 21:52Sample: 1986 2004In eluded observati ons: 19VariableCoefficie ntStd. Errort-

23、StatisticProb.C-4417.766681.1678-6.4855770.0000X70.0315280.0045076.9949430.0000R-squared0.742148Mean depe ndent var345.5232Adjusted R-squared0.726980S.D.dependent var139.7117S.E. of regressi on73.00119Akaike info criteri on11.51813Sum squared resid90595.96Schwarz criteri on11.61754Log likelihood-107

24、.4222F-statistic48.92923Durbi n- Watson stat0.572651Prob(F-statistic)0.000002表8 y对x7的回归结果Depe ndent Variable: YMethod: Least SquaresDate: 06/08/07Time: 21:52Sample: 1986 2004In cluded observati ons: 19VariableCoefficie ntStd. Errort-StatisticProb.C140.162528.966164.8388350.0002X80.1198270.0145438.23

25、95030.0000R-squared0.799739Mean depe ndent var345.5232Adjusted R-squared0.787959S.D.dependent var139.7117S.E. of regressi on64.33424Akaike info criteri on11.26536Sum squared resid70361.21Schwarz criteri on11.36478Log likelihood-105.0209F-statistic67.88941Durbi n- Watson stat0.203711Prob(F-statistic)

26、0.000000表9 y对x8的回归结果综合比较表39的回归结果，发现加入x3的回归结果最好。以x3为基础顺次加入其他解释变量，进行二元回归，具体的回归结果如下表1015所示:Depe ndent Variable: YMethod: Least SquaresDate: 06/08/07 Time: 21:53Sample: 1986 2004In cluded observati ons: 19VariableCoefficie ntStd. Errort-StatisticProb.C-754.4481149.1701-5.0576370.0001X321.788651.93268911

27、.273750.0000X213.450708.0127451.6786630.1126R-squared0.929787Mean depe ndent var345.5232Adjusted R-squared0.921010S.D.dependent var139.7117S.E. of regressi on39.26619Akaike info criteri on10.32254Sum squared resid24669.34Schwarz criteri on10.47167Log likelihood-95.06417F-statistic105.9385Durbi n- Wa

28、tson stat0.595954Prob(F-statistic)0.000000表10加入x2的回归结果Depe ndent Variable: YMethod: Least SquaresDate: 06/08/07Time: 21:53Sample: 1986 2004In cluded observati ons: 19VariableCoefficie ntStd. Errort-StatisticProb.C-508.678175.73220-6.7168020.0000X317.882003.7521214.7658370.0002X41.7533513.8443050.456

29、0900.6545R-squared0.918481Mean depe ndent var345.5232Adjusted R-squared0.908291S.D.dependent var139.7117S.E. of regressi on42.30965Akaike info criteri on10.47185Sum squared resid28641.71Schwarz criteri on10.62097Log likelihood-96.48254F-statistic90.13613Durbi n- Watson stat0.596359Prob(F-statistic)0

30、.000000表11加入x4的回归结果Depe ndent Variable: YMethod: Least SquaresDate: 06/08/07Time: 21:54Sample: 1986 2004In cluded observati ons: 19VariableCoefficie ntStd. Errort-StatisticProb.C-498.155067.21844-7.4109860.0000X323.975163.9671836.0433700.0000X5-4.3205663.553466-1.2158740.2417R-squared0.924405Mean de

31、pe ndent var345.5232Adjusted R-squared0.914956S.D.dependent var139.7117S.E. of regressi on40.74312Akaike info criteri on10.39639Sum squared resid26560.02Schwarz criteri on10.54551Log likelihood-95.76570F-statistic97.82772Durbi n- Watson stat0.607882Prob(F-statistic)0.000000表12 加入x5的回归结果Depe ndent Va

32、riable: YMethod: Least SquaresDate: 06/08/07Time: 21:54Sample: 1986 2004In cluded observati ons: 19VariableCoefficie ntStd. Errort-StatisticProb.C-1600.965346.9265-4.6147090.0003X329.937683.5347538.4695280.0000X69.9801353.1841763.1342910.0064R-squared0.948835Mean depe ndent var345.5232Adjusted R-squ

33、ared0.942440S.D.dependent var139.7117S.E. of regressi on33.51927Akaike info eriteri on10.00606Sum squared resid17976.66Schwarz eriteri on10.15518Log likelihood-92.05754F-statistie148.3576Durbi n- Watson stat1.125188Prob(F-statistie)0.000000表13加入x6的回归结果Depe ndent Variable: YMethod: Least SquaresDate:

34、 06/08/07Time: 21:54Sample: 1986 2004In eluded observati ons: 19VariableCoeffieie ntStd. Errort-StatistieProb.C-2153.028327.1248-6.5816730.0000X314.404971.35835510.604720.0000X70.0122680.0024475.0140150.0001R-squared0.967884Mean depe ndent var345.5232Adjusted R-squared0.963869S.D.dependent var139.71

35、17S.E. of regressi on26.55648Akaike info eriteri on9.540364Sum squared resid11283.94Sehwarz eriteri on9.689485Log likelihood-87.63345F-statistie241.0961Durbi n- Watson stat0.690413Prob(F-statistie)0.000000表14加入x7的回归结果Depe ndent Variable: YMethod: Least SquaresDate: 06/08/07Time: 21:54Sample: 1986 20

36、04In eluded observati ons: 19VariableCoeffieie ntStd. Errort-StatistieProb.C-400.5635103.0301-3.8878320.0013X315.542712.9163585.3294930.0001X80.0292330.0192331.5199290.1480R-squared0.927840Mean depe ndent var345.5232Adjusted R-squared0.918820S.D.dependent var139.7117S.E. of regressi on39.80687Akaike

37、 info eriteri on10.34990Sum squared resid25353.40Sehwarz eriteri on10.49902Log likelihood-95.32401F-statistie102.8643Durbi n- Watson stat0.559772Prob(F-statistie)0.000000表15加入x8的回归结果综合表1015所示，加入x7的模型的R最大，以x3、x7为基础顺次加入其他解释变量，进行三元回归，具体回归结果如下表 1620所示:Depe ndent Variable: YMethod: Least SquaresDate: 06/

38、08/07 Time: 21:55Sample: 1986 2004In cluded observati ons: 19VariableCoefficie ntStd. Errort-StatisticProb.C-2133.921340.6965-6.2634060.0000X314.960232.0946457.1421340.0000X70.0118430.0027864.2509080.0007X22.1952436.1704030.3557700.7270R-squared0.968153Mean depe ndent var345.5232Adjusted R-squared0.

39、961783S.D.dependent var139.7117S.E. of regressi on27.31242Akaike info criteri on9.637224Sum squared resid11189.52Schwarz criteri on9.836053Log likelihood-87.55363F-statistic151.9988Durbi n- Watson stat0.712258Prob(F-statistic)0.000000表16加入x2的回归结果Depe ndent Variable: YMethod: Least SquaresDate: 06/08

40、/07Time: 21:55Sample: 1986 2004In cluded observati ons: 19VariableCoefficie ntStd. Errort-StatisticProb.C-2226.420353.4425-6.2992430.0000X315.667292.4431136.4128390.0000X70.0127030.0025894.9063730.0002X4-1.6013622.553294-0.6271750.5400R-squared0.968705Mean depe ndent var345.5232Adjusted R-squared0.9

41、62445S.D.dependent var139.7117S.E. of regressi on27.07472Akaike info criteri on9.619741Sum squared resid10995.60Schwarz criteri on9.818571Log likelihood-87.38754F-statistic154.7677Durbi n- Watson stat0.704178Prob(F-statistic)0.000000表17加入x4的回归结果Depe ndent Variable: YMethod: Least SquaresDate: 06/08/

42、07Time: 21:55Sample: 1986 2004In cluded observati ons: 19VariableCoefficie ntStd. Errort-StatisticProb.C-2110.381306.2690-6.8906130.0000X318.601562.6173817.1069370.0000X70.0121390.0022855.3116650.0001X5-3.9648782.163262-1.8328230.0868R-squared0.973760Mean depe ndent var345.5232Adjusted R-squared0.96

43、8512S.D.dependent var139.7117S.E. of regressi on24.79152Akaike info criteri on9.443544Sum squared resid9219.289Schwarz criteri on9.642373Log likelihood-85.71367F-statistic185.5507Durbi n- Watson stat0.733972Prob(F-statistic)0.000000表18加入x5的回归结果Depe ndent Variable: YMethod: Least SquaresDate: 06/08/0

44、7 Time: 21:55Sample: 1986 2004In cluded observati ons: 19VariableCoefficie ntStd. Errort-StatisticProb.C-2418.859323.7240-7.4719790.0000X320.998873.3971206.1813740.0000X70.0099200.0024953.9766600.0012X65.3591842.5719502.0837050.0547R-squared0.975093Mean depe ndent var345.5232Adjusted R-squared0.9701

45、12S.D.dependent var139.7117S.E. of regressi on24.15359Akaike info criteri on9.391407Sum squared resid8750.940Schwarz criteri on9.590236Log likelihood-85.21837F-statistic195.7489Durbi n- Watson stat1.084023Prob(F-statistic)0.000000表19加入x6的回归结果Depe ndent Variable: YMethod: Least SquaresDate: 06/08/07

46、Time: 21:56Sample: 1986 2004In cluded observati ons: 19VariableCoefficie ntStd. Errort-StatisticProb.C-2013.355361.8657-5.5638180.0001X313.015782.0324206.4040780.0000X70.0116150.0025584.5403220.0004X80.0123750.0134160.9224010.3709R-squared0.969608Mean depe ndent var345.5232Adjusted R-squared0.963529

47、S.D.dependent var139.7117S.E. of regressi on26.68115Akaike info criteri on9.590455Sum squared resid10678.26Schwarz criteri on9.789285Log likelihood-87.10933F-statistic159.5158Durbi n- Watson stat0.672264Prob(F-statistie)0.000000表20 加入x8的回归结果综合上述表1620的回归结果所示，其中加入x6的回归结果最好，以 x3 x6x7为基础一次加入其他解释变量，作四元回归

48、估计，估计结果如表2124所示：Depe ndent Variable: YMethod: Least SquaresDate: 06/08/07Time: 21:57Sample: 1986 2004In eluded observati ons: 19VariableCoeffieie ntStd. Errort-StatistieProb.C-2405.108339.7396-7.0792690.0000X321.268503.6997875.7485730.0001X65.3105432.6655691.9922730.0662X70.0096890.0027663.5033860.0035X21.3026055.6553900.2303300.8212R-squared0.975187Mean depe ndent var345.5232Adjusted R-squared0.968098S.D.dependent var139.7117S.E. of regressi on24.95411Akaike info