房地產(chǎn)影響因素分析

1、房地產(chǎn)影響因素分析（背景） 2002 年以來(lái)，我國(guó)商品房銷售額大幅攀升地產(chǎn)開發(fā)和城市基礎(chǔ)設(shè)施投資的新一輪高速增長(zhǎng)。通過產(chǎn)業(yè)鏈的傳遞，進(jìn)而又拉動(dòng)鋼材、有色金屬、建材、石化等生產(chǎn)資料價(jià)格的快速上漲，刺激這些生產(chǎn)資料部門產(chǎn)能投資的成倍擴(kuò)張， 最后導(dǎo)致全社會(huì)固定資產(chǎn)投資規(guī)模過大、 增速過快情況的出現(xiàn)。 房?jī)r(jià)過快上漲在推動(dòng)投資增長(zhǎng)過快的同時(shí)，已經(jīng)成為抑制消費(fèi)的重要因素。房地產(chǎn)價(jià)格本身呈自然上漲趨勢(shì)，房?jī)r(jià)中長(zhǎng)期趨勢(shì)總是看漲。隨著我國(guó)經(jīng)濟(jì)發(fā)展，居民可支配收入提高，民間資金雄厚，大量資金需要尋找投資渠道， 而股票市場(chǎng)等投資渠道目前又處于低迷狀態(tài)，這是房地產(chǎn)投資需求不斷擴(kuò)大的經(jīng)濟(jì)背景。強(qiáng)勁的 CPI 上漲說(shuō)明當(dāng)

2、前的房?jī)r(jià)上漲并非孤立， 是有其宏觀經(jīng)濟(jì)背景的。 宏觀調(diào)控能否有效防止局部行業(yè)過熱出現(xiàn)反彈， 其中的關(guān)鍵就是要繼續(xù)加強(qiáng)和完善對(duì)房地產(chǎn)業(yè)的調(diào)控。（引言）國(guó)際上關(guān)于房地產(chǎn)有一種普遍的觀點(diǎn)： 人均收入超過 1000 美元，房地產(chǎn)市場(chǎng)呈現(xiàn)高速發(fā)展階段。 歐美等發(fā)達(dá)國(guó)家基本都經(jīng)歷了這樣一個(gè)階段。我們這篇論文，主要探討房地產(chǎn)影響因素分析，主要從人均收入對(duì)房地產(chǎn)長(zhǎng)期發(fā)展的影響闡述。年份X1X2X3Y19902551.7361510.16222704.331919911111.2361700.6233.3786.19351992590.59982026.6253.4994.655519932897.019257

3、7.4294.21291.45619943532.4713496.2367.81408.63919953983.0814282.95429.61590.86319964071.1814838.9467.41806.39919973527.5365160.3481.91997.16119982966.0575425.14792062.56919992818.8055854472.82052.620002674.2646279.98476.62111.61720012830.6886859.6479.92169.71920022906.167702.8475.122504

4、248472.2479.42359.49920043441.629421.6495.22713.878X1=建材成本（元 /平方米 ）X2=居民人均收入（元）X3= 物價(jià)指數(shù)Y= 房地產(chǎn)價(jià)格（元 /平方米）初定模型 ： Y=c+a1*x1 +a2*x2 +a3*x3+etDependent Variable: YMethod: Least SquaresDate: 06/05/05Time: 23:04Sample: 1990 2004Included observations: 15VariableCoefficieStd. Errort-StatisticProb.ntX32.537578

5、0.5904224.2979080.0013X20.1464950.0209686.9865680.0000X1-0.018010.035019-0.5144470.61716C33.20929118.27470.2807810.7841R-squared0.983094Mean1753.31dependent var7Adjusted0.978483S.D. dependent600.953R-squaredvar6S.E. of88.15143Akaike info12.0191regressioncriterion7Sum squared85477.42Schwarz12.2079res

6、idcriterion8Log likelihood-86.1437F-statistic213.21866Durbin-Watson1.504263Prob(F-statistic)0.00000stat0一：多元線性回歸Dependent Variable: YMethod: Least SquaresDate: 06/05/05Time: 23:05Sample: 1990 2004Included observations: 15VariableCoefficieStd. Errort-StatisticProb.ntX10.3360100.1510842.2239990.0445C7

7、92.0169453.44601.7466620.1043R-squared0.275612Mean1753.31dependent var7Adjusted0.219889S.D. dependent600.953R-squaredvar6S.E. of530.7855Akaike info15.5101regressioncriterion6Sum squared3662533.Schwarz15.6045residcriterion7Log likelihood-114.326F-statistic4.9461721Durbin-Watson0.275870Prob(F-statisti

8、c)0.04449stat0Dependent Variable: YMethod: Least SquaresDate: 06/05/05Time: 23:09Sample: 1990 2004Included observations: 15VariableCoefficieStd. Errort-StatisticProb.ntX35.5017790.52507510.478090.0000C-486.860220.1227-2.2117690.04555R-squared0.894128Mean1753.31dependent var7Adjusted0.885984S.D. depe

9、ndent600.953R-squaredvar6S.E. of202.9191Akaike info13.5870regressioncriterion6Sum squared535290.2Schwarz13.6814residcriterion6Log likelihood-99.9029F-statistic109.79033Durbin-Watson0.440527Prob(F-statistic)0.00000stat0Dependent Variable: YMethod: Least SquaresDate: 06/05/05Time: 23:10Sample: 1990 20

10、04Included observations: 15VariableCoefficieStd. Errort-StatisticProb.ntX20.2363470.01587914.884170.0000C561.997588.563336.3457130.0000R-squared0.944572Mean1753.31dependent var7Adjusted0.940308S.D. dependent600.953R-squaredvar6S.E. of146.8243Akaike info12.9399regressioncriterion2Sum squared280245.9S

11、chwarz13.0343residcriterion2Log likelihood-95.0493F-statistic221.53874Durbin-Watson0.475648Prob(F-statistic)0.00000stat0Dependent Variable: YMethod: Least SquaresDate: 06/07/05Time: 21:42Sample: 1990 2004Included observations: 15VariableCoefficieStd. Errort-StatisticProb.ntX32.3558330.4583405.139923

12、0.0002X20.1500860.0191577.8347140.0000C37.56794114.29910.3286810.7481R-squared0.982687Mean1753.31dependent var7Adjusted0.979802S.D. dependent600.953R-squaredvar6S.E. of85.40783Akaike info11.9096regressioncriterion1Sum squared87533.98Schwarz12.0512residcriterion2Log likelihood-86.3220F-statistic340.5

13、6479Durbin-Watson1.408298Prob(F-statistic)0.00000stat0得到結(jié)果發(fā)現(xiàn)，的系數(shù)小，然后對(duì)與回歸可決系數(shù)小，相關(guān)性差，剔出這個(gè)因素。因?yàn)閮r(jià)格更多取決于供需關(guān)系。修正之后為： Y c+a2*x2+a3*x3 et二：多重線性分析：三個(gè)表如上：X2 與 X3 存在多重共線性，1.0000000.8760730.8760731.000000Dependent Variable: YMethod: Least SquaresDate: 06/05/05Time: 23:09Sample: 1990 2004Included observations: 1

14、5VariableCoefficieStd. Errort-StatisticProb.ntX35.5017790.52507510.478090.0000C-486.860220.1227-2.2117690.04555R-squared0.894128Mean1753.31dependent var7Adjusted0.885984S.D. dependent600.953R-squaredvar6S.E. of202.9191Akaike info13.5870regressioncriterion6Sum squared535290.2Schwarz13.6814residcriter

15、ion6Log likelihood-99.9029F-statistic109.79033Durbin-Watson0.440527Prob(F-statistic)0.00000stat0Sample: 1990 2004Included observations: 15VariableCoefficieStd. Errort-StatisticProb.ntX20.2363470.01587914.884170.0000C561.997588.563336.3457130.0000R-squared0.944572Mean1753.31dependent var7Adjusted0.94

16、0308S.D. dependent600.953R-squaredvar6S.E. of146.8243Akaike info12.9399regressioncriterion2Sum squared280245.9Schwarz13.0343residcriterion2Log likelihood-95.0493F-statistic221.53874Durbin-Watson0.475648Prob(F-statistic)0.00000stat0由于引入物價(jià)指數(shù)改善小，所以模型僅一步改進(jìn)為：Y c+a2*x2 et三：異方差檢驗(yàn)：ARCH Test:F-statistic1.315

17、031Probability0.335173Obs*R-squared3.963227Probability0.265462Test Equation:Dependent Variable: RESID2Method: Least SquaresDate: 06/05/05Time: 23:46Sample(adjusted): 1993 2004Included observations: 12 after adjusting endpointsVariableCoefficieStd. Errort-StatisticProb.ntC22737.9410296.612.2082950.05

18、82RESID2(-1)0.2419520.3831440.6314930.5453RESID2(-2)-0.327760.404787-0.8097340.44159RESID2(-3)-0.273720.378355-0.7234490.49000R-squared0.330269Mean16705.2dependent var3Adjusted0.079120S.D. dependent18205.3R-squaredvar3S.E. of17470.29Akaike info22.6355regressioncriterion9Sum squared2.44E+0Schwarz22.7

19、972resid9 criterion3Log likelihood-131.813F-statistic1.3150361Durbin-Watson1.842435Prob(F-statistic)0.33517stat3臨界值所以無(wú)異方差White Heteroskedasticity Test:F-statistic0.159291Probability0.854522Obs*R-squared0.387928Probability0.823687Test Equation:Dependent Variable: RESID2Method: Least SquaresDate: 06/0

20、5/05Time: 23:46Sample: 1990 2004Included observations: 15VariableCoefficieStd. Errort-StatisticProb.ntC31063.2822612.201.3737400.1946X2-5.055759.640127-0.5244490.60954X220.0004210.0009070.4646050.6505R-squared0.025862Mean18683.0dependent var6Adjusted-0.13649S.D. dependent18673.1R-squared4 var3S.E. o

21、f19906.77Akaike info22.8123regressioncriterion6Sum squared4.76E+0Schwarz22.9539resid9 criterion7Log likelihood-168.092F-statistic0.1592971Durbin-Watson1.357657Prob(F-statistic)0.85452stat2臨界值無(wú)異方差。四：自相關(guān)分析：查表的 存在自相關(guān)廣義差分法修正： Dependent Variable: DYMethod: Least SquaresDate: 06/06/05Time: 00:18Sample(adj

22、usted): 1991 2004Included observations: 14 after adjusting endpointsVariableCoefficieStd. Errort-StatisticProb.ntDX20.1820860.0349185.2146550.0002C236.558963.273883.7386500.0028R-squared0.693820Mean544.162dependent var0Adjusted0.668305S.D. dependent148.713R-squaredvar3S.E. of85.64840Akaike info11.86

23、99regressioncriterion4Sum squared88027.77Schwarz11.9612residcriterion4Log likelihood-81.0895F-statistic27.192693Durbin-Watson1.584278Prob(F-statistic)0.00021stat7得出：回歸后可決系數(shù)降低，考慮其他方法。1迭代法：表：發(fā)現(xiàn)可決系數(shù)提高，統(tǒng)計(jì)量提高，已經(jīng)無(wú)自相關(guān)。結(jié)論： Y bY （） * （） a2*（ x2 b*x2 （）+et由下表的 b=0.681C=561.9975a2=0.236347179.2772Y*=Y 0.681Y （

24、）X*=x20.681*x2 （）Y*=179.2272 +0.2363X*+etMethod: Least SquaresDate: 06/07/05Time: 20:57Sample(adjusted): 1991 2004Included observations: 14 after adjusting endpointsVariableCoefficieStd. Errort-StatisticProb.ntE20.6805090.1776963.8296240.0024C11.6877324.888250.4696080.6471R-squared0.549989Mean15.327

25、6dependent var4Adjusted0.512488S.D. dependent133.275R-squaredvar1S.E. of93.05539Akaike info12.0358regressioncriterion3Sum squared103911.7Schwarz12.1271residcriterion2Log likelihood-82.2508F-statistic14.666012Durbin-Watson1.313042Prob(F-statistic)0.00239stat72改進(jìn)模型方程（對(duì)數(shù)法，然后用迭代法） ：L L （）（）（ L L（）可決系數(shù)很高

26、， F 統(tǒng)計(jì)量相對(duì) 1 中也有提高， DW=1.81>1.361 無(wú)自相關(guān)。Dependent Variable: LYMethod: Least SquaresDate: 06/06/05Time: 10:24Sample(adjusted): 1991 2004Included observations: 14 after adjusting endpointsConvergence achieved after 7 iterationsVariableCoefficieStd. Errort-StatisticntProb.LX20.5862030.1002435.8477990.0001C2.5258100.8823502.8625940.0154AR(1)0.5671440.2204572.5725890.0259R-squared0.980054Mean7.46009dependent var6Adjusted0.976428S.D. dependent0.35133R-squaredvar1S.E. of0.053941Akaike info-2.8144regressioncriterion4