*---------------------------------*
* GARCH Model Fit *
*---------------------------------*Conditional Variance Dynamics
There are difference between ret( SNP) and re1t(NSEI)
----------------------------------- GARCH Model : apARCH(1,1) Mean Model : ARFIMA(1,0,0) Distribution : std Optimal Parameters ------------------------------------ Estimate Std. Error t value Pr(>|t|) mu 0.000036 0.000247 0.14737 0.882840 ar1 -0.048548 0.015318 -3.16936 0.001528 omega 0.000004 0.000002 2.36223 0.018165 alpha1 0.061980 0.009038 6.85804 0.000000 beta1 0.929079 0.012530 74.14894 0.000000 gamma1 0.067265 0.047143 1.42682 0.153631 delta 2.029480 0.032782 61.90755 0.000000 shape 5.851426 0.518903 11.27654 0.000000 Robust Standard Errors: Estimate Std. Error t value Pr(>|t|) mu 0.000036 0.000259 0.14057 0.888208 ar1 -0.048548 0.013808 -3.51589 0.000438 omega 0.000004 0.000002 1.88772 0.059063 alpha1 0.061980 0.015451 4.01136 0.000060 beta1 0.929079 0.021014 44.21271 0.000000 gamma1 0.067265 0.056776 1.18474 0.236122 delta 2.029480 0.169875 11.94690 0.000000 shape 5.851426 0.525345 11.13825 0.000000 LogLikelihood : 10531.7 Information Criteria ------------------------------------ Akaike -5.0425 Bayes -5.0304 Shibata -5.0425 Hannan-Quinn -5.0382
Weighted Ljung-Box Test on Standardized Residuals
------------------------------------ statistic p-value Lag[1] 0.6906 0.4060 Lag[2*(p+q)+(p+q)-1][2] 1.1329 0.6582 Lag[4*(p+q)+(p+q)-1][5] 2.6692 0.5188 d.o.f=1 H0 : No serial correlation Weighted Ljung-Box Test on Standardized Squared Residuals ------------------------------------ statistic p-value Lag[1] 0.2302 0.63135 Lag[2*(p+q)+(p+q)-1][5] 5.2651 0.13324 Lag[4*(p+q)+(p+q)-1][9] 9.1664 0.07512 d.o.f=2 Weighted ARCH LM Tests ------------------------------------ Statistic Shape Scale P-Value ARCH Lag[3] 1.953 0.500 2.000 0.16227 ARCH Lag[5] 7.602 1.440 1.667 0.02500 ARCH Lag[7] 9.344 2.315 1.543 0.02609 Nyblom stability test ------------------------------------ Joint Statistic: 131.3225 Individual Statistics: mu 0.4580 ar1 0.1933 omega 0.8532 alpha1 0.5161 beta1 0.4291 gamma1 0.0970 delta 0.4220 shape 0.2261 Asymptotic Critical Values (10% 5% 1%) Joint Statistic: 1.89 2.11 2.59 Individual Statistic: 0.35 0.47 0.75 Sign Bias Test ------------------------------------ t-value prob sig Sign Bias 0.65217 0.5143 Negative Sign Bias 0.02349 0.9813 Positive Sign Bias 0.37243 0.7096 Joint Effect 1.46310 0.6908 Adjusted Pearson Goodness-of-Fit Test: ------------------------------------ group statistic p-value(g-1) 1 20 11.45 0.90767 2 30 33.88 0.24378 3 40 45.17 0.22993 4 50 63.76 0.07654 Elapsed time : 2.417678 > re1t.aparch.t = ugarchfit(ret1, spec=arma.aparch.t) > show(re1t.aparch.t) *---------------------------------* * GARCH Model Fit * *---------------------------------* Conditional Variance Dynamics ----------------------------------- GARCH Model : apARCH(1,1) Mean Model : ARFIMA(1,0,0) Distribution : std Optimal Parameters ------------------------------------ Estimate Std. Error t value Pr(>|t|) mu 0.000347 0.000174 1.9939 0.046167 ar1 0.069182 0.017870 3.8714 0.000108 omega 0.000030 0.000026 1.1749 0.240044 alpha1 0.081367 0.010825 7.5162 0.000000 beta1 0.910433 0.010036 90.7161 0.000000 gamma1 0.641660 0.094727 6.7738 0.000000 delta 1.407281 0.184595 7.6236 0.000000 shape 7.942371 1.009616 7.8667 0.000000 Robust Standard Errors: Estimate Std. Error t value Pr(>|t|) mu 0.000347 0.000175 1.98721 0.046900 ar1 0.069182 0.015231 4.54211 0.000006 omega 0.000030 0.000037 0.83308 0.404798 alpha1 0.081367 0.011118 7.31842 0.000000 beta1 0.910433 0.013455 67.66450 0.000000 gamma1 0.641660 0.146132 4.39096 0.000011 delta 1.407281 0.265997 5.29059 0.000000 shape 7.942371 1.114134 7.12874 0.000000 LogLikelihood : 9788.523 Information Criteria ------------------------------------ Akaike -6.2020 Bayes -6.1866 Shibata -6.2020 Hannan-Quinn -6.1965 Weighted Ljung-Box Test on Standardized Residuals ------------------------------------ statistic p-value Lag[1] 0.4772 0.4897 Lag[2*(p+q)+(p+q)-1][2] 0.4789 0.9691 Lag[4*(p+q)+(p+q)-1][5] 0.5468 0.9915 d.o.f=1 H0 : No serial correlation Weighted Ljung-Box Test on Standardized Squared Residuals ------------------------------------ statistic p-value Lag[1] 0.715 0.3978 Lag[2*(p+q)+(p+q)-1][5] 3.862 0.2718 Lag[4*(p+q)+(p+q)-1][9] 6.972 0.2015 d.o.f=2 Weighted ARCH LM Tests ------------------------------------ Statistic Shape Scale P-Value ARCH Lag[3] 0.009886 0.500 2.000 0.92080 ARCH Lag[5] 6.080525 1.440 1.667 0.05750 ARCH Lag[7] 7.741753 2.315 1.543 0.05985 Nyblom stability test ------------------------------------ Joint Statistic: 3.4347 Individual Statistics: mu 0.20446 ar1 0.10876 omega 1.15284 alpha1 1.57680 beta1 1.58706 gamma1 0.07561 delta 1.40797 shape 0.14866 Asymptotic Critical Values (10% 5% 1%) Joint Statistic: 1.89 2.11 2.59 Individual Statistic: 0.35 0.47 0.75 Sign Bias Test ------------------------------------ t-value prob sig Sign Bias 1.6905 0.09103 * Negative Sign Bias 1.1973 0.23129 Positive Sign Bias 0.6159 0.53802 Joint Effect 3.0967 0.37695 Adjusted Pearson Goodness-of-Fit Test: ------------------------------------ group statistic p-value(g-1) 1 20 21.24 0.3235 2 30 36.33 0.1642 3 40 42.93 0.3065 4 50 52.28 0.3479 Elapsed time : 2.504623
0 Comments