question related to fixed parameters in "rugarch" package

Dear all:

I am trying to learn how to use "rugarch" package for ARIMA + sGARCH modeling. When I tried to fix some parameters (statistically insignificant coefficients), I got the following output:

*---------------------------------*
*          GARCH Model Fit        *
*---------------------------------*

Conditional Variance Dynamics
-----------------------------------
GARCH Model     : sGARCH(1,1)
Mean Model      : ARFIMA(4,0,0)
Distribution    : std

Optimal Parameters
------------------------------------
         Estimate  Std. Error  t value Pr(>|t|)
ar1     0.000000          NA       NA       NA
ar2     0.630141    0.024459  25.7627  0.0e+00
ar3     0.085549          NA       NA       NA
ar4    -0.244984    0.022802 -10.7439  0.0e+00
omega   0.004524    0.001050   4.3097  1.6e-05
alpha1  0.266026    0.045341   5.8672  0.0e+00
beta1   0.569310    0.070464   8.0795  0.0e+00
shape   7.382851    0.898399   8.2178  0.0e+00

Robust Standard Errors:
         Estimate  Std. Error  t value Pr(>|t|)
ar1     0.000000          NA       NA       NA
ar2     0.630141    0.028354  22.2244 0.000000
ar3     0.085549          NA       NA       NA
ar4    -0.244984    0.027248  -8.9908 0.000000
omega   0.004524    0.001524   2.9696 0.002982
alpha1  0.266026    0.060709   4.3820 0.000012
beta1   0.569310    0.107354   5.3031 0.000000
shape   7.382851    1.443335   5.1151 0.000000

LogLikelihood : 1165.098

Information Criteria
------------------------------------

And here is the code:


spec<- ugarchspec(variance.model = list(model = "sGARCH", garchOrder = c(1,1)), mean.model = list(armaOrder = c(4, 0), include.mean = FALSE), distribution.model = "std", start.pars = startPar, fixed.pars = list(ar1 = 0,ar3 = 0))fit<- ugarchfit(data = myData, spec = spec1, solver = "nlminb")

Is it true that the coefficients for ar1 and ar3 should be 0.0 or set to NA?

Your help will be appreciated.


John Li
(224) 372-7082
	[[alternative HTML version deleted]]