《VAR模型应用案例(完成)》由会员分享，可在线阅读，更多相关《VAR模型应用案例(完成)（12页珍藏版）》请在金锄头文库上搜索。
1、VAR模型应用实例众所周知，经济的发展运行离不开大量能源的消耗，尤其是在现代经济发展的过程中，能源的重要性日益提升。我国自改革开放以来，经济发展取得长足的进步，经济增长率一直 处于较高的速度，经济的高速增长带来了能源的大量消耗，进而带来了我国能源生产的巨大提高。因此，研究经济增长率与能源生产增长率之间的关系具有重要的意义，能为生源生产提供一定的指导意义。1.基本的数据我们截取1978 2015年中国经济增长速度（GDP增速）和中国能源生产增长速度数据， 具体数据如下：表1 佃78 2016年中国经济和能源生产增长率年份国内生产总值增长速度（%）能源生产增长速度（%）年份国内生产总值增长速度（%
2、）能源生产增长速度（%）197811.710.419979.20.319797.63.719987.8-2.719807.8-1.319997.71.619815.1-0.820008.55198295.620018.36.4198310.86.720029.16198415.29.220031014.1198513.49.9200410.115.619868.93200511.411.1198711.73.6200612.76.9198811.25200714.27.919894.26.120089.7519903.92.220099.43.119919.30.9201010.69.1199
3、214.22.320119.59199313.93.620127.93.21994136.920137.82.21995118.720147.30.919969.93.120156.91.22.序列平稳性检验（单位根检验）使用Eviews9.0来创建一个无约束的 VAR模型，用gdp表示的是中国经济的增长率，用 nysc表示中国能源生产的增长率，下面分别对gdp和nysc进行单位根检验，验证序列是否平稳，能否达到建立 VAR模型的建模前提。bd Senes: GDP Workfile: UNTITLED:Untitled口 | 回 |ViewProtObjertProprtiPrintMm色F
4、reertSampleGt nrSheet GraphAugmented Dickey-Fuller Unit Root Test on GDPNull Hyp(rthesis： GDP has a unit rootExogenous: ConstantLag Length: 3 Autorriaiic-cased on SIC, maxlag二g)Statistic Prob?Augm的柜U Dick即亠Full盯馆st stM待tic胡酣5迢06056Test critical values： 1% level-3.6394075% level-2.951125前临I凹创-2614300
5、* MacKinnon (1996) one-sid&d p*values.Augmented Dickey-Fuller Test Equation Dopendent Variath D(GDF)Method: Least SquaresDate: 05/17/17 Time: 10:55Sample (adjusted): 1382 2015Included observations: M after adjustmentsVariableCoefficientstd. Errort-statisticProb.GDP (-1)-0S5&17102211143 867553 ODOSDG
6、DP(-1)0 625B310,1935293.2327550.0031D；GDP-2)Q.D492400.1755170 200544D7311D(GDPg)0.2645370.1673431 5S31450.1242C6 5400502.2229613 0417450.0006R-squared0458475Mean depend ntvar0.052941Adjusted R-squ a redH r 車 鼻i0,3837624 nn-nn-mS.D.dependentvar2&45731图2.1经济增速（GDP）的单位根检验0 SeHesiWorlcfile: UNTlTLED:Unt
7、itled卜口| ViewProcObjectPropertiesPrintNameFreezeSa m p kGenrSheetGraphALgniented Dickey-Fuller Unrt Root Test ora NY SCNull H /potriesis: FJYSC has a unit rootExogenous: ConstantLag Length: 1 (Automatic-based on SIC, maxlag=9)t-Statistic Prob.*Dickw卜FullwrtEst statistic9了5987QO045Test critical value
8、s:1)level-3 5257845% level-2 945S4210% level-2,511M1MacKinnon (1996)one-sided valuesAugmented Dickey-Fuller Test EquationDependent Variable: D(NYSC)Method: Least SquaresDate: 05/17/17 Time: 10:58Sample (adjusted): 1980 2015Included observations: after adjustmentsVariableCoefficient Sid. Error t-Stal
9、istic Prob.NYSCM)-0.5309060.1349D5-39359870.00S4-D(NVSCL-D)0.4385490 1500552.9225850.00S2C2 7469380.85726632043300.0030R*squaredA34306SMean dependent var4.069444Adjusted R-squared0 303254S.D. dependent var3.510704S.E. cf regression5 93C431Aka ike info criterion5.0C7S31Sum squared resici233.3351Schwa
10、rz criteridn5J99791Log likelihood-68.22096Hannan-Quinn critef.5,113839F-statistic8616745Durbin-Watsor stat1.950251Pro bF-5tati stic)0 000975图2.2能源生产增速（nysC的单位根检验经过检验，在1%的显著性水平上，gdp和nysc两个时间序列都是平稳的，符合建模的 条件，我们建立一个无约束的VAR模型。3.VAR模型的估计ViewPro;ObjedPtirrtNameFreezeEstimateForca stVector Autoregression E
11、stimates/ector Autoreg郵ion EstimatesDate 05/17/17 Time: 1103Sample (adjusred: 1&80 2015Indudediobservaticns. 36 after adjustments Standard errors In ()4t-statisties in GDPNYSCGDP(-1)0.92&544(0.164995 003690 271590(0.23599) 1.15096GDF-2)-0.530405(0.1662d-3.19096-0.292356(0.23730)-1.22942NYSCC-1)-0 05
12、2225 也1伽) -0.451560 846356(P.16&43)5.11612NYSC(-2)o.iesioo tOJ1349) (1.639771-0 357569(0.16234)-2.20253C6.184518(1.508874.105392.363291(2.15B27)1.3 26 &61Rsquared&.492S650.554397Adj. R-squared0 4270990.496639Sum sq. res；ids130.5151267.0S33SE equation2.0518692.534956F-statistic7.5228909.t41791Log lik
13、elihood-74.265257.15117Akaike AIC440362& 119500Schwarz SC4.623&5S5 339442Mean dependent9 7388995.C166&7S D dependent2.7108544.137805Determinant resid covariance (dof adj)JO. 72390Determinant resid covariance2278215Log likelihood-1584312Altaie information criterion9.357287Schwarz cr it&ri on9.797154图3.1模型的估计结果回Van VAR01 Workfile-：九曲酣”住蟆型:;Untitl巳或I ViewProcObjectPrintNameF r&tze 1Estimate | Forecast5tatsImpLlIseReside1 u乍gtimatien. Froc：LS 1 2 GDP HlfSCGDF -C（bl）*HT（-l）C（LZ）*（H）P（-2）CnSC（-l） + C（b 4）*HKC（-2）+
