可化为线性的非线性回归模型估计受约束回归检验

《可化为线性的非线性回归模型估计受约束回归检验》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、实验三可化为线性的非线性回归模型估计、受约束回归检验一、实验目的：(1)掌握可化为线性的的非线性回归模型的估计方法(2)掌握Chow检验的基本原理和主要用途(3)掌握Chow分割点检验和Chow预测检验的操作过程，判断分割点。二、实验要求做可化为线性的非线性回归模型估计，掌握Chow稳定性检验三、实验原理普通最小二乘法、模型参数线性受约束检验法、Chow检验法四、预备知识普通最小二乘法原理、t检验、F检验、Chow检验五、实验步骤13..建立工作文件并录入全部数据obsY

2、KLobsYKLI13722.700

3、3078.220113.000021442.5201684.43067.0000031752.3702742.77084.0000041451.2901973.82027.0000055149.3005917.010327.000062291.1601758770120.000071345.170939.100058.000008656.7700694.940031.000009370.1800363.480016.00000101590.3602511.99066.0000011616.7100973.73

4、0058.0000012617.9400516.010028.00000134429.1903785.91061.00000145749.0208688.030254.0000151781.3702798.90083.00000161243.0701808.44033.0000017812.70001118.81043.00000181899.7002052.16061.00000193692.8506113.110240.0000204732.9009228.250222.0000212180.230286

5、6.65080.00000222539.7602545.63096.00000233046.9504787.900222.0000242192.6303255.290163.0000255364.8308129.680244.0000264834.6805260.200145.0000277549.5807518.790138.000028867.9100984.520046.00000294611.39018626.94218.000030170.3000610.910019.0000031325.5300

6、1523.19045.00000■1.设定并估计可化为线性的非线性回归模型ln(y)=c+aln(k)+bln(l)+ueviews结果VariableCoefficientStd.Errort-StatisticProb.C1.1539940.7276111.5860040.1240LOG(K)0.6092360.1763783.4541490.0018LOG(L)0.3607960.2015911.7897410.0843R-squared0.809925Meandependentvar7.493997A

7、djustedR-squared0.796348S-D.dependentvar0.942960S.E.ofregression0.425538Akaikeinfocriterion1.220839Sumsquaredresid5.070303Schwarzcriterion1.359612Loglikelihood•15.92300F-statistic59.65501Durbin-Watsonstat0793209Prob(F-statistic)0.000000DependentVariable:LOG

8、(Y)Method:LeastSquaresDate:10/17/12Time:12:57Sample:131Includedobservations:31根据数据得到模型的估计结果为：Ln(y)=1.153994+0.609236In(k)+0.360796In(I)(1.586004)(3.454149)(1.789741)RAR=0.809925R_AR_=0.796348D.W.=0.793209elAel+...+eiAei=5.070303F=59.65501df=随机干扰项的方差估计值为：6=5

9、.070303/(31-3)=0.181082我们根据得出的回归结果表明：(1)(2)In(y)变化的81.90%可由其他两个变量的变化来解释，在5%的显著注水平下。F统计量的临界值为F0.05(2,28)=3.34,表明模型的线性关系显著成立，(3)自由度为n-k-l=28的t检验量临界值为t0.025(28)=2.048,因此，ln(y)的参数显著的异与零，却不拒绝ln(k)与ln(l)前参数为