
摘要:运用E views 软件建立我国农村居民全年人均消费性支出的计量经济模型,对影响我国农村居民全年人均消费性支出的可能因素进行分析,发现农村居民全年人均纯收入、农村居民消费价格指数、人均实际纯收入、人均实际消费性支出对我国农村居民全年人均消费性支出具有重要的影响
关键字:农村居民 人均消费性支出 影响因素 多元线性回归
一、问题的提出
今年以来,全国上下认真贯彻落实科学发展观,以农业增产、农民增收为目的,加大各项惠农措施落实力度,多措并举做好农村劳动力转移就业工作,克服金融危机和严重干旱等自然灾害带来的不利影响,使全市农村经济保持了稳定发展的良好态势,农民现金收入持续增长,生活消费水平继续提高。
我国是一个农业大国,至今仍有9亿农村人口,占全国人口总数的70%,农民是我国最大的群体,农村消费能力的提升直接关系到国民经济的全局。从农村市场看,中国有近六成人口生活在农村。农村城镇化的进程对经济增长的带动作用是非常明显的,世界上还没有哪个国家有规模如此巨大的城镇化。农村居民的收入虽然低于城市居民,但是基数巨大,且农村人口的收入也在稳定增长。
随着经济的发展,我国农民的收入水平和消费水平的结构也发生了很大变化,农民生活水平的提高和消费的增加对于实现国民经济又好又快发展、正确处理好内需和外需的关系至关重要。但从总体来看,农民消费水平仍然较低,调查显示有的地区都不及城市居民人均消费支出的三分之一。而且消费结构不合理,局限于食品类等生存基本需求品,消费在衣着装饰等方面的极少。而影响农民消费水平的根本原因是农民的收入。
农民生活消费支出主要包括食品、衣着、医疗卫生、教育文化、家庭设备、交通等方面,本文只挑选了五种典型的消费支出作为代表来分析农村居民的消费结构。
二、理论依据
(一)、
1.
三、模型设定
(一)、影响因素的分析
1、.农民收入偏低,增收困难抑制消费。消费的多少在很大程度上决定于可支配收入,虽然农村居民近几年人均收入增长明显,但相对仍处于较低水平。
2、为下一代消费支出过大影响其他消费。一是教育费用支出比重过大。调查显示,高中、大学教育负担沉重严重影响其他消费支出。
3、.医疗支出不确定性,使农村居民不敢轻易消费。农村生活水平近几年有大幅提升,在解决了基本温饱问题之后,越来越多的农村居民对医疗养老增加了关注度,对医疗保险的投入度也较以往有所增加,特别是在新农村合作医疗保险开办之后。
4、主动负债消费理念还未普及,社会需求潜能被隐藏。改革开放后,我国居民长期保持着量人为出的消费理念,特别是对于普通农村居民长期以来形成了勤俭持家的习惯,消费观念较为保守,提倡“量人为出、知足常乐”的消费观念
(二)、影响因素的选择
影响农民人均生活消费的因素有很多。经分析有如下:变量选择和说明:被解释变量即自变量:农民人均生活消费支出;解释变量即因变量:农民人均收入,农民人均食品消费支出,衣着消费支出,农民人均交通和通讯消费支出,农民人均医疗保健消费支出。
(三)、模型形式的设计
为此设定了如下对数形式的计量经济模型:
Y = β1+β2Xt+β3X2t+β4X3t+β5X4t+μt
其中
Y-----农民人均生活消费支出
X-----农民人均收入
X1-----农民人均食品消费支出
X2-----衣着消费支出
X3----农民人均交通和通讯消费支出
X4----农民人均医疗保健消费支出
μt ---- 随机干扰
四、数据的收集
(一)、全国各地区农村基本情况—人均消费情况 (2011)单位:元
| 指标 | 农村居民家庭人均收入 | 农村居民家庭平均每人生活消费支出 | ||||
| 食品 | 衣着 
  | 交通和通讯 | 医疗保健 | |||
| 全国合计 | 6977.3 | 5221.1 | 2107.3 | 341.3 | 547 | 436.8 | 
| 北京市 | 14735.7 | 11077.7 | 3593.5 | 862.6 | 1228.2 | 1035.2 | 
| 天津市 | 12321.2 | 6725.4 | 2376 | 611.7 | 781.6 | 571.7 | 
| 河北省 | 7119.7 | 4711.2 | 1579.7 | 334.1 | 520.2 | 434.7 | 
| 山西省 | 5601.4 | 4587 | 1729.9 | 401.9 | 458.8 | 349.3 | 
| 内蒙古自治区 | 61.6 | 5507.7 | 2067 | 395.2 | 728.9 | 534.2 | 
| 辽宁省 | 8296.5 | 5406.4 | 2116.3 | 446.1 | 577.7 | 482.9 | 
| 吉林省 | 7510.0 | 5305.8 | 1872.1 | 397.5 | 5.3 | 673.6 | 
| 黑龙江省 | 7590.7 | 5333.6 | 2072.4 | 473.8 | 576.3 | 573.6 | 
| 上海市 | 16053.8 | 11049.3 | 4517.2 | 4.5 | 1308.9 | 908.6 | 
| 江苏省 | 10805.0 | 8094.6 | 2839.9 | 554.8 | 923.9 | 5.6 | 
| 浙江省 | 13070.7 | 9965.1 | 3714.8 | 717.5 | 1380.6 | 921.3 | 
| 安徽省 | 6232.2 | 4957.3 | 2055.2 | 297 | 475.2 | 440.5 | 
| 福建省 | 8778.6 | 6540.9 | 3032.2 | 395.4 | 728.5 | 321.2 | 
| 江西省 | 61.6 | 4659.9 | 2106.3 | 233.6 | 393.3 | 346.7 | 
| 山东省 | 8342.1 | 5900.6 | 2107.1 | 399.8 | 753.1 | 508.4 | 
| 河南省 | 6604.0 | 4320 | 1559.7 | 362.8 | 427.9 | 399.7 | 
| 湖北省 | 67.9 | 5010.7 | 1954.6 | 272.1 | 414.4 | 438.2 | 
| 湖南省 | 6567.1 | 5179.4 | 2343.1 | 260.4 | 421.7 | 396.5 | 
| 广东省 | 9371.7 | 6725.6 | 3301.1 | 277.3 | 682.5 | 398.5 | 
| 广西壮族自治区 | 5231.3 | 4210.9 | 1844.9 | 123.9 | 384.8 | 301.3 | 
| 海南省 | 46.0 | 4166.1 | 2137.9 | 139.8 | 370.3 | 290.1 | 
| 重庆市 | 80.4 | 4502.1 | 2108.6 | 309 | 401.7 | 375.3 | 
| 四川省 | 6128.6 | 4675.5 | 2161.7 | 281.9 | 431.1 | 413.1 | 
| 贵州省 | 4145.4 | 3455.8 | 16.5 | 186.2 | 304.5 | 246.3 | 
| 云南省 | 4722.0 | 3999.9 | 1884 | 209.1 | 393 | 309.3 | 
| 自治区 | 4904.3 | 2741.6 | 1384.7 | 331.2 | 348.9 | 65.8 | 
| 陕西省 | 5027.9 | 4491.7 | 1345 | 285.4 | 406.7 | 533.4 | 
| 甘肃省 | 3909.4 | 36.9 | 1548.2 | 246.7 | 366.6 | 339.3 | 
| 青海省 | 4608.5 | 4536.8 | 1716.4 | 347.5 | 450.9 | 308.1 | 
| 宁夏回族自治区 | 5410.0 | 4726.6 | 1762.5 | 380 | 483.4 | 444.7 | 
| 维吾尔自治区 | 5442.2 | 4397.8 | 15.5 | 372.1 | 530.6 | 376.9 | 
| 指 标 | 1990 | 1995 | 2000 | 2010 | 2011 | |
| 调查户数 (户) | 66960.00 | 67340.00 | 68116.00 | 68190.00 | 73630.00 | |
| 调查户人口 (人) | ||||||
| 平均每户常住人口 | 4.80 | 4.48 | 4.20 | 3.95 | 3.90 | |
| 平均每户整半劳动力 | 2.92 | 2.88 | 2.76 | 2.85 | 2.78 | |
| 平均每个劳动力负 | ||||||
| 担人口(含本人) | 1. | 1.56 | 1.52 | 1.39 | 1.40 | |
| 平均每人年收入(元) | ||||||
| 总收入 | 990.38 | 2337.87 | 3146.21 | 8119.51 | 9833.14 | |
| 工资性收入 | 138.80 | 353.70 | 702.30 | 2431.05 | 2963.43 | |
| 家庭经营收入 | 815.79 | 1877.42 | 2251.28 | 4937.48 | 5939.79 | |
| 财产性收入 | 35.79 | 40.98 | 45.04 | 202.25 | 228.57 | |
| 转移性收入 | 65.77 | 147.59 | 548.74 | 701.35 | ||
| 现金收入 | 676.67 | 1595.56 | 2381.60 | 7088.76 | 8638.51 | |
| 工资性收入 | 136.43 | 352.88 | 700.41 | 2427. | 2959.74 | |
| 家庭经营收入 | 481.19 | 1116.73 | 1498.81 | 3955.36 | 4810.37 | |
| 财产性收入 | 59.05 | 38.19 | 38. | 168.33 | 185.76 | |
| 转移性收入 | 87.76 | 143.49 | 537.18 | 682. | ||
| 纯收入 | 686.31 | 1577.74 | 2253.42 | 5919.01 | 6977.29 | |
| 工资性收入 | 138.80 | 353.70 | 702.30 | 2431.05 | 2963.43 | |
| 家庭经营纯收入 | 518.55 | 1125.79 | 1427.27 | 2832.80 | 3221.98 | |
| 财产性收入 | 28.96  | 
| 40.98 | 45.04 | 202.25 | 228.57 | |||
| 转移性收入 | 57.27 | 78.81 | 452.92 | 563.32 | ||
| 平均每人年支出 (元) | ||||||
| 总支出 | 903.47 | 2138.33 | 2652.42 | 6991.79 | 81.63 | |
| 家庭经营费用支出 | 241.09 | 621.71 | 654.27 | 1915.62 | 2431.05 | |
| 购置生产性固定资产 | 20.29 | 62.33 | 63.90 | 193.26 | 265.75 | |
| 税费支出 | 38.66 | 88.65 | 95.52 | 8.57 | 11.67 | |
| 消费支出 | 584.63 | 1310.36 | 1670.13 | 4381.82 | 5221.13 | |
| 财产性支出 | 18.80 | 55.28 | 19.74 | 49.25 | 12.27 | |
| 转移性支出 | 148.86 | 443.27 | 699.76 | |||
| 现金支出 | 639.06 | 1545.81 | 2140.37 | 6307.43 | 7984.94 | |
| 家庭经营费用支出 | 162.90 | 454.74 | 544.49 | 1757.58 | 2269.19 | |
| 购买生产性固定资产 | 20.46 | 62.32 | 63.91 | 193.26 | 265.75 | |
| 税费支出 | 33.37 | 76.96 | .81 | 8.56 | 11.65 | |
| 消费支出 | 374.74 | 859.43 | 1284.74 | 3859.33 | 4733.35 | |
| 财产性支出 | 47.59 | 92.35 | 9.82 | 49.25 | 12.27 | |
| 转移性支出 | 147.60 | 439.45 | 692.73 | |||
(一)、模型估计
1、农村家庭总收入单线图,农村家庭总收入逐年增加。(X-农村家庭总收入 Y-年份)
表1
2、利用E views软件,输入Y、X、X2、X3、X4、X5等数据,采用这些数据对模型进行OLS回归,结果如表:
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/14/14 Time: 22:01 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | -136.9270 | 154.0293 | -0.8867 | 0.3822 | 
| X | 0.024548 | 0.041835 | 0.586771 | 0.5624 | 
| X1 | 1.065002 | 0.145190 | 7.335242 | 0.0000 | 
| X2 | 1.791296 | 0.604166 | 2.9908 | 0.00 | 
| X3 | 1.503512 | 0.179686 | 8.367460 | 0.0000 | 
| X4 | 1.584193 | 0.484330 | 3.270900 | 0.0030 | 
| R-squared | 0.992351 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.990880 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 190.5708 | Akaike info criterion | 13.50529 | |
| Sum squared resid | 944248.1 | Schwarz criterion | 13.78011 | |
| Log likelihood | -210.0846 | Hannan-Quinn criter. | 13.59638 | |
| F-statistic | 674.6076 | Durbin-Watson stat | 1.933440 | |
| Prob(F-statistic) | 0.000000 | |||
由此可见,该模型R²=0.9924, F=674.608
则,我国农村居民全年人均消费性支出模型的估计式为:
Y= -136.927+0.02455 Xt +1.065 X1t+1.7913X2t +1.5035 X3t+1.5842 X4t +μt
(二)、模型检验
1、经济意义检验。
模型估计结果说明:农村居民全年人均纯收入、农村居民消费价格指数、人均实际消费性支出的增加都将带来我国农村居民全年人均消费性支出的增加,与理论分析和经验判断一致。该模型通过了经济意义上的检验,系数符号均符合经济意义。
2、统计意义检验。
R2=0.9924说明模型的拟合优度较好,F=674.608符合F检验,因而农民人均收入、农民人均食品消费支出、衣着消费支出、农民人均交通和通讯消费支出、农民人均医疗保健消费支出五个解释变量对农村居民全年人均消费性支出的99.2%的离差做出解释,且解释变量联合起来对被解释变量有显著影响。
3、多重共线性的检验
X 、X1、 X2 、X3 、X4的相关系数如表:
| X | X1 | X2 | X3 | X4 | |
| X | 1 | 0.7944 | 0.7450 | 0.7459 | 0.7258 | 
| X1 | 0.7944 | 1 | 0.6536 | 0.7450 | 0.7705 | 
| X2 | 0.7450 | 0.6536 | 1 | 0.7858 | 0.7830 | 
| X3 | 0.7459 | 0.7450 | 0.7858 | 1 | 0.7500 | 
| X4 | 0.7258 | 0.7705 | 0.7830 | 0.8500 | 1 | 
通过简单相关系数检验法:由表2可知任意两个解释变量之间的零阶相关系数<0.8。由此知该模型不存在多重共线性
用Y 分别对X、X2、X3、X4、X5作一元线性回归,结果如图:
图2-1
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:20 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | 735.9259 | 290.9029 | 2.529799 | 0.0169 | 
| X | 0.637597 | 0.036258 | 17.58476 | 0.0000 | 
| R-squared | 0.911563 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.908615 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 603.2415 | Akaike info criterion | 15.70297 | |
| Sum squared resid | 10917008 | Schwarz criterion | 15.79458 | |
| Log likelihood | -249.2476 | Hannan-Quinn criter. | 15.73334 | |
| F-statistic | 309.2239 | Durbin-Watson stat | 1.913390 | |
| Prob(F-statistic) | 0.000000 | |||
图2-2
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:19 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | -55.22996 | 445.3505 | -0.124015 | 0.9021 | 
| X1 | 2.531038 | 0.193050 | 13.11076 | 0.0000 | 
| R-squared | 0.851406 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.8453 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 781.9424 | Akaike info criterion | 16.22190 | |
| Sum squared resid | 18343018 | Schwarz criterion | 16.31351 | |
| Log likelihood | -257.5504 | Hannan-Quinn criter. | 16.25227 | |
| F-statistic | 171.21 | Durbin-Watson stat | 1.155038 | |
| Prob(F-statistic) | 0.000000 | |||
由此可见,该模型R²=0.8514
图2-3
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:18 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | 1606.443 | 476.0873 | 3.374262 | 0.0021 | 
| X2 | 10.424 | 1.176877 | 8.1533 | 0.0000 | 
| R-squared | 0.724920 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.715751 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 1063.905 | Akaike info criterion | 16.83774 | |
| Sum squared resid | 33956823 | Schwarz criterion | 16.92935 | |
| Log likelihood | -267.4039 | Hannan-Quinn criter. | 16.86811 | |
| F-statistic | 79.05937 | Durbin-Watson stat | 1.422873 | |
| Prob(F-statistic) | 0.000000 | |||
由此可见,该模型R²=0.7249
图2-4
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:16 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | 700.9945 | 425.0283 | 1.92 | 0.1095 | 
| X3 | 4.746216 | 0.394249 | 12.03863 | 0.0000 | 
| R-squared | 0.828502 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.822785 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 840.0476 | Akaike info criterion | 16.36526 | |
| Sum squared resid | 21170400 | Schwarz criterion | 16.45686 | |
| Log likelihood | -259.8441 | Hannan-Quinn criter. | 16.39562 | |
| F-statistic | 144.9287 | Durbin-Watson stat | 1.400608 | |
| Prob(F-statistic) | 0.000000 | |||
图2-5
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:15 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | 1416.370 | 220.4140 | 6.425955 | 0.0000 | 
| X4 | 6.955591 | 0.340809 | 20.40905 | 0.0000 | 
| R-squared | 0.932815 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.930576 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 525.7865 | Akaike info criterion | 15.42813 | |
| Sum squared resid | 8293544. | Schwarz criterion | 15.51974 | |
| Log likelihood | -244.8501 | Hannan-Quinn criter. | 15.45849 | |
| F-statistic | 416.5292 | Durbin-Watson stat | 1.863327 | |
| Prob(F-statistic) | 0.000000 | |||
由图2-1、2-2、2-3、2-4、2-5知X4的R²最大
所以以以yx4作为基础 再用Y分别对XX4、X1X4、 X2 X4 、X3 X4作线性回归;结果如图
图3-1
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:32 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | 960.5453 | 206.4310 | 4.653106 | 0.0001 | 
| X | 0.283188 | 0.066834 | 4.237155 | 0.0002 | 
| X4 | 4.128208 | 0.720749 | 5.727660 | 0.0000 | 
| R-squared | 0.958504 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.9553 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 420.2775 | Akaike info criterion | 15.00877 | |
| Sum squared resid | 5122363. | Schwarz criterion | 15.14618 | |
| Log likelihood | -237.1403 | Hannan-Quinn criter. | 15.05432 | |
| F-statistic | 334.9350 | Durbin-Watson stat | 2.190815 | |
| Prob(F-statistic) | 0.000000 | |||
图3-2
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:33 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | 625.1550 | 245.5205 | 2.546244 | 0.0165 | 
| X1 | 0.927956 | 0.205668 | 4.511925 | 0.0001 | 
| X4 | 4.834639 | 0.539972 | 8.953493 | 0.0000 | 
| R-squared | 0.960526 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.957803 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 409.9148 | Akaike info criterion | 14.95884 | |
| Sum squared resid | 4872875. | Schwarz criterion | 15.09625 | |
| Log likelihood | -236.3414 | Hannan-Quinn criter. | 15.00438 | |
| F-statistic | 352.8259 | Durbin-Watson stat | 2.3169 | |
| Prob(F-statistic) | 0.000000 | |||
图3-3
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:33 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | 1421.820 | 240.0823 | 5.922218 | 0.0000 | 
| X2 | -0.079905 | 1.260403 | -0.063396 | 0.9499 | 
| X4 | 6.996936 | 0.738555 | 9.473813 | 0.0000 | 
| R-squared | 0.932824 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.928192 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 534.7379 | Akaike info criterion | 15.49049 | |
| Sum squared resid | 8292394. | Schwarz criterion | 15.62790 | |
| Log likelihood | -244.8478 | Hannan-Quinn criter. | 15.53604 | |
| F-statistic | 201.3524 | Durbin-Watson stat | 1.872625 | |
| Prob(F-statistic) | 0.000000 | |||
图3-4
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:33 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | 877.1122 | 205.5988 | 4.266134 | 0.0002 | 
| X3 | 1.676713 | 0.360801 | 4.7201 | 0.0001 | 
| X4 | 4.986980 | 0.498313 | 10.00772 | 0.0000 | 
| R-squared | 0.961492 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.958836 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 404.87 | Akaike info criterion | 14.93404 | |
| Sum squared resid | 4753548. | Schwarz criterion | 15.07146 | |
| Log likelihood | -235.9447 | Hannan-Quinn criter. | 14.97959 | |
| F-statistic | 362.0467 | Durbin-Watson stat | 1.414208 | |
| Prob(F-statistic) | 0.000000 | |||
由图可知R²=0.9615最大,则以YX3 X4为基础 用Y分别对X X3 X4、X1 X3 X4、 X2 X3 X4作多元线性回归 结果如图
图4-1
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:38 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | 7.4620 | 179.5072 | 3.606886 | 0.0012 | 
| X | 0.213345 | 0.055493 | 3.844521 | 0.0006 | 
| X3 | 1.323020 | 0.310980 | 4.254357 | 0.0002 | 
| X4 | 3.2721 | 0.606033 | 5.399357 | 0.0000 | 
| R-squared | 0.974796 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.972096 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 333.3395 | Akaike info criterion | 14.57267 | |
| Sum squared resid | 3111227. | Schwarz criterion | 14.755 | |
| Log likelihood | -229.1627 | Hannan-Quinn criter. | 14.63340 | |
| F-statistic | 360.9839 | Durbin-Watson stat | 1.470526 | |
| Prob(F-statistic) | 0.000000 | |||
图4-2
| Dependent Variable: Y | |||||
| Method: Least Squares | |||||
| Date: 06/15/14 Time: 15:38 | |||||
| Sample (adjusted): 1 32 | |||||
| Included observations: 32 after adjustments | |||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | |
| C | 110.9190 | 151.0029 | 0.734549 | 0.4687 | |
| X1 | 0.910258 | 0.114721 | 7.934563 | 0.0000 | |
| X3 | 1.5832 | 0.2037 | 8.077163 | 0.0000 | |
| X4 | 2.942736 | 0.381508 | 7.713436 | 0.0000 | |
| R-squared | 0.988146 | Mean dependent var | 5495.281 | ||
| Adjusted R-squared | 0.986876 | S.D. dependent var | 1995.508 | ||
| S.E. of regression | 228.6073 | Akaike info criterion | 13.81836 | ||
| Sum squared resid | 1463317. | Schwarz criterion | 14.00157 | ||
| Log likelihood | -217.0937 | Hannan-Quinn criter. | 13.87909 | ||
| F-statistic | 778.0155 | Durbin-Watson stat | 1.865370 | ||
| Prob(F-statistic) | 0.000000 | ||||
图4-3
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:39 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | 914.0456 | 213.0030 | 4.291234 | 0.0002 | 
| X2 | -0.725666 | 0.971530 | -0.746931 | 0.4613 | 
| X3 | 1.715755 | 0.367321 | 4.671000 | 0.0001 | 
| X4 | 5.316622 | 0.668529 | 7.952714 | 0.0000 | 
| R-squared | 0.962244 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.958199 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 407.9865 | Akaike info criterion | 14.97681 | |
| Sum squared resid | 4660683. | Schwarz criterion | 15.16003 | |
| Log likelihood | -235.6290 | Hannan-Quinn criter. | 15.03754 | |
| F-statistic | 237.8710 | Durbin-Watson stat | 1.582733 | |
| Prob(F-statistic) | 0.000000 | |||
由图可知R²=0.9881最大,则以YX1X3 X4为基础 用Y分别对XX1 X3 X4、X1 X2X3 X4作多元线性回归 结果如图
图5-1
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:41 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | 125.6046 | 143.06 | 0.877946 | 0.3877 | 
| X | 0.085475 | 0.041365 | 2.066373 | 0.0485 | 
| X1 | 0.783504 | 0.1246 | 6.283661 | 0.0000 | 
| X3 | 1.508428 | 0.203960 | 7.395717 | 0.0000 | 
| X4 | 2.540381 | 0.410174 | 6.193431 | 0.0000 | 
| R-squared | 0.9765 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.988248 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 216.3246 | Akaike info criterion | 13.73404 | |
| Sum squared resid | 1263501. | Schwarz criterion | 13.96306 | |
| Log likelihood | -214.7446 | Hannan-Quinn criter. | 13.80995 | |
| F-statistic | 652.7227 | Durbin-Watson stat | 1.966234 | |
| Prob(F-statistic) | 0.000000 | |||
图5-2
| Dependent Variable: Y | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:42 | ||||
| Sample (adjusted): 1 32 | ||||
| Included observations: 32 after adjustments | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | -165.81 | 144.2608 | -1.148255 | 0.2609 | 
| X2 | 1.965431 | 0.519826 | 3.780939 | 0.0008 | 
| X1 | 1.119986 | 0.109547 | 10.22383 | 0.0000 | 
| X3 | 1.532974 | 0.170420 | 8.995275 | 0.0000 | 
| X4 | 1.5713 | 0.478330 | 3.300887 | 0.0027 | 
| R-squared | 0.992249 | Mean dependent var | 5495.281 | |
| Adjusted R-squared | 0.991101 | S.D. dependent var | 1995.508 | |
| S.E. of regression | 188.2426 | Akaike info criterion | 13.45594 | |
| Sum squared resid | 956752.1 | Schwarz criterion | 13.68496 | |
| Log likelihood | -210.2951 | Hannan-Quinn criter. | 13.53185 | |
| F-statistic | 8.1597 | Durbin-Watson stat | 1.886714 | |
| Prob(F-statistic) | 0.000000 | |||
White检验结果如下:
| F-statistic | 3.562028 | Prob. F(20,11) | 0.0173 | |
| Obs*R-squared | 27.71987 | Prob. Chi-Square(20) | 0.1162 | |
| Scaled explained SS | 20.21128 | Prob. Chi-Square(20) | 0.4448 | |
| Test Equation: | ||||
| Dependent Variable: RESID^2 | ||||
| Method: Least Squares | ||||
| Date: 06/15/14 Time: 15:59 | ||||
| Sample: 4 35 | ||||
| Included observations: 32 | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. | 
| C | -6294.1 | 234087.0 | -2.6018 | 0.0211 | 
| X | -120.5027 | 61.84256 | -1.948540 | 0.0773 | 
| X^2 | 0.004849 | 0.010672 | 0.454373 | 0.6584 | 
| X*X1 | 0.040398 | 0.053073 | 0.761176 | 0.4626 | 
| X*X2 | 0.110758 | 0.218160 | 0.507690 | 0.6217 | 
| X*X3 | -0.025956 | 0.067862 | -0.382484 | 0.7094 | 
| X*X4 | -0.102732 | 0.2076 | -0.494153 | 0.6309 | 
| X1 | 881.4267 | 357.9961 | 2.462113 | 0.0316 | 
| X1^2 | -0.183045 | 0.131816 | -1.388635 | 0.1924 | 
| X1*X2 | -3.166096 | 0.815849 | -3.880740 | 0.0026 | 
| X1*X3 | -0.347085 | 0.237409 | -1.461969 | 0.1717 | 
| X1*X4 | 1.826167 | 0.700216 | 2.608006 | 0.0243 | 
| X2 | 4813.793 | 1054.928 | 4.563148 | 0.0008 | 
| X2^2 | -9.460861 | 1.743218 | -5.427240 | 0.0002 | 
| X2*X3 | -3.169044 | 0.945138 | -3.352994 | 0.00 | 
| X2*X4 | 19.441 | 3.484934 | 5.578768 | 0.0002 | 
| X3 | 95.26270 | 167.2194 | 0.569687 | 0.5803 | 
| X3^2 | 0.524129 | 0.221240 | 2.369049 | 0.0372 | 
| X3*X4 | 1.684955 | 1.259445 | 1.337854 | 0.2079 | 
| X4 | -27.149 | 760.8548 | -3.665810 | 0.0037 | 
| X4^2 | -7.5541 | 1.373581 | -5.499612 | 0.0002 | 
| R-squared | 0.866246 | Mean dependent var | 29507.75 | |
| Adjusted R-squared | 0.623057 | S.D. dependent var | 44557.74 | |
| S.E. of regression | 27356.53 | Akaike info criterion | 23.51596 | |
| Sum squared resid | 8.23E+09 | Schwarz criterion | 24.47785 | |
| Log likelihood | -355.2553 | Hannan-Quinn criter. | 23.83480 | |
| F-statistic | 3.562028 | Durbin-Watson stat | 1.928746 | |
| Prob(F-statistic) | 0.017282 | |||
由表3可以看出,nR²=27.71987,由White检验知,在α=0.05下,查χ²分布表,得临界值=30.1435,比较计算的χ²统计量与临界值,因为nR²=27.71987<30.1435,表明模型不存在异方差。
5、自相关的检验
通过DW检验法 由表1知该模型的DW统计量=1.9334 查DW分布表可得临界值dL=1.144 dU=1.808。因为dU=1.808 在实际应用中,农民消费支出方面有很多,通过线性回归模型也可以较为准确的判断今后的农民消费情况。在现实生活中,所得预测结果不可能与生活完全一致,但是对增进农民收入、改变农民消费结构有很大的意义。   可以看出,我国农民的费结构,基本上还是在食品、医疗等生活必需品上消费较多,而花在衣着装饰上的较少,但比起过去农民在家庭设备上的支出有了明显提高。而制约农民消费的关键还是农民收入不足。 因此,国家应该调整相应的农业,切实增加农民收入,增强消费的经济基础,通过增加消费拉动经济增长,通过经济增长带动消费的增加。此外还应培育农村居民正确的消费观念,要加快形成积极的消费观念,在生产发展的基础上努力提高生活质量,使生活更加富有意义;要克服“只知道买价格低、便宜的商品,养儿防身防老”等片面观念。 七、意见 【参考文献】 于勇 曲敏《农村消费市场培育及国际经验借鉴》 (一)、.拓宽农民增收渠道,增加农民收入。一是加快推进农业现代化进程,实现农业生产标准化和商品化,降低农民生产经营成本;二是提高农民在粮食生产至最终零售整个过程中的收入分配份额,如大力发展农村粮食经济合作组织,提高粮食生产效率及农民在粮食收购过程中的议价能力;三是进一步规范和培育农村粮食流通主体,培育科学发达的农产品流通体系。 (二)、加强财政支持“三农”力度,稳步推进农业发展农民增收。发展中国现代化农业,提升农民生活水平,必须加强支持。一是加大种粮补贴力度,粮食直补、农资综合补贴、良种补贴和柴油补贴,根据生产资料价格涨幅等进一步扩大补贴;二是建立和完善农业生产风险防范和保险机制,推动农业保险发展,扩大承保范围,最大限度地降低农民风险;三是提高财政对“三农”的支出比重,向现代农业、农村教育和卫生等社会事业、农村基础设施建设倾斜。 (三)、完善农村社会保障体系,减轻农民消费后顾之忧。一是加大资金投入。统筹经济社会发展水平、农村居民基本生活需求、消费者物价指数、财政承受能力等因素,努力增加农村社会保障资金投入,坚持事权财权统一,更加科学地划分事权财权,合理确定和地方的保障责任。加强对农村社会保障经费管理与监督,确保专款专用。二是加强协调协作。增进各部门互联、互动、互补,增强推进合力,确保工作责任到位、落实到位、措施保障到位。加强各项农村社会保障制度之间的衔接,加强农村社会保障与就业再就业、农村扶贫开发等之间的配套,形成梯次保障结构,提高整体保障效能。 (四)、发展农村合理化消费信贷,鼓励农民消费。利用现有的银行信贷登记咨询系统,建立个人信用评估和查询系统,减少信贷风险;通过市场机制引入高素质信贷人员,针对农村基层富裕程度不一,消费需求各异的情况,应采取差异性营销策略,开发不同层次的消费信贷业务。金融机构要深入农村,针对农民需求特点,从贷款项目、方式、利率、期限等方面开发出适应农村特点的消费信贷新品种;要适当降低消费贷款利率,减轻农民的利息负担,据调查现阶段农民申请贷款,主要在农村信用社,他们的贷款利率一般上浮100%~130%之间,大大高于同期城市贷款水平,建议对资信较好的农民采取类于小额担保贷款利率水平的予以发放,促进农村消费信贷的发展;要尽快改变消费贷款手续繁琐、附加条件多、门槛高的现状,简化贷款手续,方便农户贷款。 (五)、加快农村基础设施建设,改善农村消费环境。一是加强农村水、电、路、通讯等设施的建设,尤其是加大电网改造力度、整顿农村电价,降低用电成本,鼓励农民对家电的消费,为空调、冰箱、洗衣机等家电和大型农机具进入农村创造条件。二是健全农村市场体系,拓宽农村流通渠道,逐步构建以城镇为中心,以专业批发为特色,以农村集贸市场为依托,合理布局的农村商业辐射网络。三是建立健全农村消费品市场物流配送体系。应采取投资参股的方式扶持,制定相应的优惠,在农村建立一批股份制商业企业,建立健全农村消费品市场物流配送体系,从而减少中间环节,使不法商贩无机可乘。
