GARCH模型的参数

《GARCH模型的参数》由会员上传分享，免费在线阅读，更多相关内容在工程资料-天天文库。

1、GARCH=C(1)+C(2)*RESID(-1)^2+C(3)*GARCH(-1)VariableCoefficientStd.Errorz-StatisticProb.  VarianceEquationC1.31E-064.16E-073.1433690.0017RESID(-1)^20.0589560.0064079.2024130.0000GARCH(-1)0.9333690.007214129.37570.0000R-squared-0.001087    Meandependentvar0.000421Adju

2、stedR-squared-0.000267    S.D.dependentvar0.012759S.E.ofregression0.012760    Akaikeinfocriterion-6.110837Sumsquaredresid0.198649    Schwarzcriterion-6.098279Loglikelihood3730.610    Hannan-Quinncriter.-6.106110Durbin-Watsonstat2.092936对GARCH（1,1）模型做出的LM检验，发现F统计量和卡

3、方统计量都大于0.05，所以不存在ARCG效应HeteroskedasticityTest:ARCHF-statistic1.628845    Prob.F(1,1217)0.2021Obs*R-squared1.629342    Prob.Chi-Square(1)0.2018下面是GARCH(1，2)和GARCH（2,1）模型阐述估计GARCH（1,2）GARCH=C(1)+C(2)*RESID(-1)^2+C(3)*GARCH(-1)+C(4)*GARCH(-2)VariableCoefficientStd.Err

4、orz-StatisticProb.  VarianceEquationC7.87E-073.87E-072.0361990.0417RESID(-1)^20.0339050.0139222.4354280.0149GARCH(-1)1.4755680.2173026.7904050.0000GARCH(-2)-0.5137710.202024-2.5431240.0110R-squared-0.001087    Meandependentvar0.000421AdjustedR-squared-0.000267    S

5、.D.dependentvar0.012759S.E.ofregression0.012760    Akaikeinfocriterion-6.113214Sumsquaredresid0.198649    Schwarzcriterion-6.096471Loglikelihood3733.061    Hannan-Quinncriter.-6.106912Durbin-Watsonstat2.092936GARCH（2,1）模型GARCH=C(1)+C(2)*RESID(-1)^2+C(3)*RESID(-2)^2

6、+C(4)*GARCH(-1)VariableCoefficientStd.Errorz-StatisticProb.  VarianceEquationC1.67E-065.31E-073.1440960.0017RESID(-1)^20.0037690.0169180.2227910.8237RESID(-2)^20.0738540.0192683.8328910.0001GARCH(-1)0.9139570.00955495.658410.0000R-squared-0.001087    Meandependentv

7、ar0.000421AdjustedR-squared-0.000267    S.D.dependentvar0.012759S.E.ofregression0.012760    Akaikeinfocriterion-6.115724Sumsquaredresid0.198649    Schwarzcriterion-6.098981Loglikelihood3734.592    Hannan-Quinncriter.-6.109422Durbin-Watsonstat2.092936EARCH模型LOG(GARC

8、H)=C(1)+C(2)*ABS(RESID(-1)/@SQRT(GARCH(-1)))+C(3)        *RESID(-1)/@SQRT(GARCH(-1))+C(4)*LOG(GARCH(-1))VariableCoefficientStd.Errorz-StatisticPr