資源描述:
《計(jì)量經(jīng)濟(jì)學(xué)實(shí)驗(yàn)四-多重共線(xiàn)性的檢驗(yàn)與修正.doc》由會(huì)員上傳分享,免費(fèi)在線(xiàn)閱讀,更多相關(guān)內(nèi)容在應(yīng)用文檔-天天文庫(kù)。
1、《計(jì)量經(jīng)濟(jì)學(xué)》實(shí)驗(yàn)報(bào)告四開(kāi)課實(shí)驗(yàn)室:財(cái)經(jīng)科學(xué)實(shí)驗(yàn)室年月日班級(jí):學(xué)號(hào):姓名:指導(dǎo)教師簽字:實(shí)驗(yàn)項(xiàng)目名稱(chēng):多重共線(xiàn)性的檢驗(yàn)與修正成績(jī):驗(yàn)證性□綜合性□設(shè)計(jì)性實(shí)驗(yàn)性質(zhì):t【實(shí)驗(yàn)?zāi)康摹空莆斩嘀毓簿€(xiàn)性的檢驗(yàn)與修正方法并能運(yùn)用Eviews軟件進(jìn)行實(shí)現(xiàn)【實(shí)驗(yàn)要求】能根據(jù)OLS的估計(jì)結(jié)果判斷是否存在多重共線(xiàn)性,熟悉逐步回歸法修正模型的基本操作步驟,讀懂各項(xiàng)上機(jī)榆出結(jié)果的含義并能進(jìn)行分析【實(shí)驗(yàn)軟件】Eviews軟件【實(shí)驗(yàn)內(nèi)容】根據(jù)給定的案例數(shù)據(jù)按實(shí)驗(yàn)要求進(jìn)行操作【實(shí)驗(yàn)方案與進(jìn)度】實(shí)驗(yàn):設(shè)蔬菜銷(xiāo)售量Y與人口(X1)、價(jià)格(X2)、糧食(X3)、收
2、入(X4)、副食(X5)和儲(chǔ)蓄(X6)等資料如下:obsYX1X2X3X4X5X619787.450425.58.1217.517.80185.8521.6819797.605422.38.3222.919.51185.3521.0819807.855418.08.3623.718.93185.1021.0319817.805419.28.2021.119.05184.8020.7319826.900384.28.8623.319.57184.6021.9319837.470372.57.7019.119.95184.2522
3、.4919847.385372.98.4618.220.89181.3523.2619857.225380.88.8822.223.27179.3024.3919868.130401.79.0027.626.06178.1025.0419878.720406.58.8028.828.55176.2525.5319889.145410.59.2627.830.12174.3526.64198910.105447.08.6224.432.78174.2527.53199010.170452.88.4424.132.21179.35
4、28.12199110.540467.19.6627.833.57173.8531.35199210.635495.29.6819.534.86179.5034.58199310.455500.011.3225.436.60166.8541.78199410.995525.012.3028.440.35158.2542.85199512.380550.012.8835.445.00155.0046.75199611.770561.014.0234.849.87141.0549.21要求:(1)將Y關(guān)于其他變量線(xiàn)性回歸Depen
5、dentVariable:YMethod:LeastSquaresDate:06/03/13Time:16:48Sample:19781996Includedobservations:19VariableCoefficientStd.Errort-StatisticProb.C-1.6.-0.0.8032X10.0.5.0.0003X2-0.0.-2.0.0172X30.0.2.0.0492X40.0.3.0.0071X50.0.0.0.5004X60.0.1.0.1148R-squared0.Meandependentvar
6、9.AdjustedR-squared0.S.D.dependentvar1.S.E.ofregression0.Akaikeinfocriterion0.Sumsquaredresid0.Schwarzcriterion0.Loglikelihood3.F-statistic142.6067Durbin-Watsonstat2.Prob(F-statistic)0.(2)方程線(xiàn)性顯著性檢驗(yàn)由(1)表中的數(shù)據(jù)可知F統(tǒng)計(jì)量的值為142.6067,查表得=3,顯然142.6067>=3,說(shuō)明方程具有線(xiàn)性顯著性。(3)第六個(gè)解釋變量
7、的變量顯著性檢驗(yàn)。在0.05的顯著水平條件下,查表得=2.179,由(1)表中數(shù)據(jù)可知,的t檢驗(yàn)值為1.,明顯小于臨界值,則接受原假設(shè),說(shuō)明對(duì)Y的影響不顯著,對(duì)方程沒(méi)有意義。(4)用直觀判斷法判斷模型是否有多重共線(xiàn)性?由(1)表中數(shù)據(jù)可知,該模型判定系數(shù)=0.,調(diào)整的判定系數(shù)=0.,數(shù)值都接近于1,解釋變量對(duì)被解釋變量的解釋程度很高,而F統(tǒng)計(jì)值為142.6067,明顯顯著。但是如果給定0.05的顯著性水平,=2.179,顯然,與系數(shù)不能通過(guò)t檢驗(yàn),而且的符號(hào)與預(yù)計(jì)的不同,這表明很可能存在嚴(yán)重的多重共線(xiàn)性。(5)對(duì)解釋變量之間的
8、相關(guān)系數(shù)進(jìn)行檢查,是否可懷疑自變量之間存在嚴(yán)重的多重共線(xiàn)性。X1X2X3X4X5X6X110.0.0.-0.0.X20.10.0.-0.0.X30.0.10.-0.0.X40.0.0.1-0.0.X5-0.-0.-0.82884-0.1-0.X60.0.0.0.-0.1由相關(guān)