《2021-2022年收藏的精品资料计量经济学论文eviews分析 房价的计量经济分析.doc》由会员分享,可在线阅读,更多相关《2021-2022年收藏的精品资料计量经济学论文eviews分析 房价的计量经济分析.doc(8页珍藏版)》请在taowenge.com淘文阁网|工程机械CAD图纸|机械工程制图|CAD装配图下载|SolidWorks_CaTia_CAD_UG_PROE_设计图分享下载上搜索。
1、房价的计量经济分析引言:近改革开放20多年来,从来没有哪一个行业像房地产业这样盛产亿万富翁,各种富豪排行榜上,房地产富豪连年占据半壁江山;“中国十大暴利行业”中,房地产业每年都是“第一名”。是什么造就了这样的状况。房地产的问题,在开发商,政府,购房者三者来看,就是一场完完全全的博弈。而这场博弈的焦点则是房价问题。如果说开发商与政府之间的博弈是围绕“土地”这个关键词,那么整个房地产市场则在价格上开展了新一轮的对峙。先是开发商与购房者在房价涨跌上僵持不下;再有开发商与政府之间的土地成本论;最后则是关于房地产是否归为暴利行业的争执,“价格”成了市场关注的焦点。而对于房价的构成因素,至今仍然是不透明的
2、。公布房价成本成为另政府极为头疼的一件事。房价成本是一个非常复杂的集合体,并且项目间差异性较大,同时还有软资产、品牌等组成部分,特别是现在的商品房,追求品质、功能完善以及个性化成本构成越来越难衡量。 写作目的:通过对一系列影响房价的基本因素的分析,了解对其主要因素和次要因素。并对这些因素进行统计推断和经济意义上的检验。选择拟和效果最好的最为结论。在一定层面上分析房地产如此暴利的因素。当然笔者的能力有限,并不能全面的分析这一问题。仅仅就几个因素进行分析。写作方法:理论分析及计量分析方法,将会用到Eviews软件进行帮助分析。关键词:房价成本 计量假设检验 最小二乘法 拟合优度 现在我们以2003
3、年的数据,选取30个省市的数据为例进行分析。在Eviews软件中选择建立截面数据。现在我们以2003年的数据,选取31个省市的数据为例进行分析。令Y=各地区建筑业总产值。(万元)X1=各地区房屋竣工面积。(万平方米)X2=各地区建筑业企业从业人员。(人)X3=各地区建筑业劳动生产率。(元/人)X4=各地区人均住宅面积。(平方米)X5=各地区人均可支配收入。(元)数据如下:YX1X3X2X4X5 12698521 4254.800 569767.0 129961.0 24.77140 13882.62 5208402. 1465.800 238957.0 147063.0 23.09570 10
4、312.91 7799313. 4748.300 989317.0 70048.00 23.16710 7239.060 5401279. 1313.300 591276.0 89151.00 22.99680 7005.030 2576575. 1450.700 265953.0 61074.00 20.05310 7012.900 10170794 3957.100 966790.0 82496.00 20.23510 7240.580 3469281. 1626.800 303837.0 77486.00 20.70590 7005.170 4401878. 2181.300 44151
5、8.0 68033.00 20.49200 6678.900 11958034 3609.200 505185.0 153910.0 29.34530 14867.49 27949354 17730.00 2727006. 100569.0 24.43530 9262.460 31272779 16183.90 2429352. 127430.0 31.02330 13179.53 6227073. 4017.600 910691.0 66407.00 20.75480 6778.030 5493441. 2952.100 553611.0 108288.0 30.29870 9999.540
6、 3593356. 2750.900 574705.0 70826.00 22.61980 6901.420 14813618 9139.800 2072530. 60728.00 24.48080 8399.910 6345217. 3433.600 932901.0 66056.00 20.20090 6926.120 8729958. 4840.800 1048763. 81761.00 22.90280 7321.980 8188402. 4969.700 1119106. 74553.00 24.42580 7674.200 15163242 8105.000 1492820. 10
7、1932.0 24.93280 12380.43 2818466. 1721.600 353700.0 77472.00 24.17320 7785.040 394053.0 121.5000 61210.00 55361.00 23.43200 7259.250 5862095. 4939.600 817997.0 69432.00 25.72440 8093.670 12253374 8784.600 2070534. 59748.00 26.35850 7041.870 2122907. 980.3000 293310.0 72152.00 18.19430 6569.230 39679
8、57. 2248.700 522470.0 69238.00 24.92940 7643.570 293427.0 121.3000 36593.00 73205.00 19.92990 8765.450 4404362. 1580.000 410311.0 93212.00 21.75050 6806.350 2236860. 1327.200 449409.0 46857.00 21.11380 6657.240 747325.0 242.9000 101501.0 61046.00 19.10550 6745.320 1080546. 578.7000 88225.00 61459.00
9、 22.25500 6530.480 3196774. 1450.800 203375.0 95835.00 20.78110 7173.540 做多重共线性检验:引入的变量太多,可能存在变量间的共线性,影响方程的估计。首先进行做多重共线性检验可以减少变量使后面的分析变得简洁。X1X2X3X4X5YX110.9608709909074460.2713751927607750.5386972790690410.4183068002953290.961473842608042X20.96087099090744610.1250293750973190.477885891518730.2798506
10、233443580.898672551511606X30.2713751927607750.12502937509731910.5408809599699260.836240848942410.467710383760092X40.5386972790690410.477885891518730.54088095996992610.686512808507740.589777148826127X50.4183068002953290.2798506233443580.836240848942410.6865128085077410.58982338526214Y0.96147384260804
11、20.8986725515116060.4677103837600920.5897771488261270.589823385262141可以看出有多重共线性。数 97数 97得的的的采取逐步回归法:第一次回归,我们可以根据T检验值和可决系数看出:X1的效果最好:Dependent Variable: YMethod: Least SquaresDate: 12/06/10 Time: 17:37Sample (adjusted): 1 31Included observations: 31 after adjustmentsVariableCoefficientStd. Errort-Sta
12、tisticProb.X11651.40387.6770318.835080.0000C903234.0502408.21.7978090.0826R-squared0.924432Mean dependent var7446408.Adjusted R-squared0.921826S.D. dependent var7227629.S.E. of regression2020815.Akaike info criterion31.93824Sum squared resid1.18E+14Schwarz criterion32.03076Log likelihood-493.0427F-s
13、tatistic354.7601Durbin-Watson stat1.930762Prob(F-statistic)0.000000而X1于X2存在严重自相关,所以引入第二个变量时将X2排除。通过比较发现引入X3时,拟合优度最大,所以加入X3Dependent Variable: YMethod: Least SquaresDate: 12/06/10 Time: 17:40Sample (adjusted): 1 31Included observations: 31 after adjustmentsVariableCoefficientStd. Errort-StatisticProb
14、.X11547.35457.8319726.756040.0000X360.575779.1368996.6297950.0000C-3711880.765709.2-4.8476370.0000R-squared0.970594Mean dependent var7446408.Adjusted R-squared0.968493S.D. dependent var7227629.S.E. of regression1282914.Akaike info criterion31.05893Sum squared resid4.61E+13Schwarz criterion31.19771Lo
15、g likelihood-478.4134F-statistic462.0886Durbin-Watson stat2.098685Prob(F-statistic)0.000000 X3与X5也存在严重共线性,在引入第三个变量时同时排除X5,那只能引入X4了Dependent Variable: YMethod: Least SquaresDate: 12/06/10 Time: 17:47Sample (adjusted): 1 31Included observations: 31 after adjustmentsVariableCoefficientStd. Errort-Stati
16、sticProb.X11569.18666.7446723.510290.0000X364.0494510.562586.0638100.0000X4-69455.16102797.7-0.6756490.5050C-2476469.1985261.-1.2474280.2230R-squared0.971083Mean dependent var7446408.Adjusted R-squared0.967870S.D. dependent var7227629.S.E. of regression1295550.Akaike info criterion31.10668Sum square
17、d resid4.53E+13Schwarz criterion31.29171Log likelihood-478.1536F-statistic302.2316Durbin-Watson stat2.298423Prob(F-statistic)0.000000但是引入后通过T检验X4不显著,同时常数项C也变得不显著,且拟合度没有显著提高。所以剔除X4。通过该检验最终模型为:Y = 1547.354325*X1 + 60.57576644*X3 - 3711880.158T= 26.75604 6.629795 -4.847637F-statistic354.7601R-squared0.
18、970594Durbin-Watson stat2.098685 以上指标都显示拟合得很好。 异方差检验White Heteroskedasticity Test:F-statistic1.742532Probability0.161697Obs*R-squared8.011602Probability0.155597Test Equation:Dependent Variable: RESID2Method: Least SquaresDate: 12/06/10 Time: 18:05Sample: 1 31Included observations: 31VariableCoeffici
19、entStd. Errort-StatisticProb.C-3.19E+124.46E+12-0.7158550.4807X11.15E+083.54E+080.3249150.7479X123913.00420466.630.1911890.8499X1*X3-756.30894598.986-0.1644510.8707X369425884952903000.7285720.4730X32-184.1939462.0769-0.3986220.6936R-squared0.258439Mean dependent var1.49E+12Adjusted R-squared0.110127
20、S.D. dependent var2.04E+12S.E. of regression1.92E+12Akaike info criterion59.58019Sum squared resid9.25E+25Schwarz criterion59.85774Log likelihood-917.4929F-statistic1.742532Durbin-Watson stat2.029951Prob(F-statistic)0.161697从结果来看应该勉强是不存在异方差的,但是同方差的概率有点小,不能让人信服。而通过残差图发现残差没有很明显的波动、X-Y的图也较符合线性关系即模型设定没多
21、大问题、且从White Heteroskedasticity Test 中各变量的系数也十分不显著不能判别残差是否与解释变量有关。没办法,只能用加权最小二乘法进行修正。异方差修正-加权最小二乘法 Dependent Variable: YMethod: Least SquaresDate: 12/06/10 Time: 18:13Sample (adjusted): 1 31Included observations: 31 after adjustmentsWeighting series: 1/ABS(RESID)VariableCoefficientStd. Errort-Statist
22、icProb.X11543.8124.266721361.82620.0000X360.882210.92521265.803540.0000C-3721097.59118.40-62.943140.0000Weighted StatisticsR-squared0.999999Mean dependent var7466651.Adjusted R-squared0.999999S.D. dependent var34381715S.E. of regression29817.20Akaike info criterion23.53532Sum squared resid2.49E+10Sc
23、hwarz criterion23.67410Log likelihood-361.7975F-statistic310479.3Durbin-Watson stat2.158638Prob(F-statistic)0.000000Unweighted StatisticsR-squared0.970589Mean dependent var7446408.Adjusted R-squared0.968489S.D. dependent var7227629.S.E. of regression1283009.Sum squared resid4.61E+13Durbin-Watson sta
24、t2.099900通过修正以后拟合度有所提高,且通过再次异方差检验通过了。自相关检验Breusch-Godfrey Serial Correlation LM Test:Obs*R-squared0.505922Probability0.776498Test Equation:Dependent Variable: RESIDMethod: Least SquaresDate: 12/06/10 Time: 18:26Presample missing value lagged residuals set to zero.VariableCoefficientStd. Errort-Stati
25、sticProb.X1-6.77803562.81436-0.1079060.9149X31.2596669.7075420.1297620.8978C-73457.01800910.8-0.0917170.9276RESID(-1)-0.1250060.210750-0.5931470.5582RESID(-2)-0.0678210.201592-0.3364250.7393R-squared0.016320Mean dependent var-2178.743Adjusted R-squared-0.135015S.D. dependent var1239503.S.E. of regre
26、ssion1320530.Akaike info criterion31.17165Sum squared resid4.53E+13Schwarz criterion31.40294Log likelihood-478.1606F-statistic0.107840Durbin-Watson stat1.862550Prob(F-statistic)0.978723从结果看自相关检验也通过,模型不存在自相关。正态性检验由检验知残差符合正态性假设。稳定性检验由图知模型十分稳定,具有很好的预测能力。综上最后的出模型为Y = 1543.81157*X1 + 60.8822121*X3 - 3721
27、097.247结论:我们总认为房产总价值与许多成分有关,其实在最后我们看到并不是这样。但现实中房价成本具有相当大的难度。不管是资金成本很难简单地以招拍挂价格进行测算,还是融资成本比较难核算。而且房地产的利润要以综合成本衡量。种种原因构成了房价成本确定的难度。而房产行业的暴利,开发商的暴利是来源于开发商的阶层优越感和特殊占有地位,而与之相对的是老百姓的阶层卑微感和相对剥削感。房地产业的暴利如果继续维持,考验的不仅是中国经济的稳定,更是老百姓忍耐的限度。而且这种房产的暴利行为导致了从2003年10月开始的通货膨胀,并造成了中国越来越大的金融风险。我国房价的公开将会采取怎么样的方式,笔者将和大家一起拭目以待。