garchfit significance

what is the exact level of t-stat in the garchfit outcome to become significance for 1% and 5%

    Parameter      Value        Error       T Statistic 

 -----------   -----------   ------------   -----------
           C    -0.0013526     0.0013368       -1.0119
           K    0.00027647     0.00014944       1.8500
    GARCH(1)    0.77871        0.10106          7.7057
     ARCH(1)    0.073557       0.028595         2.572

Answer 1

[Wayne King]

You can evaluate that statistic on the t(N-p) pdf where N is the length of the time series and p is the number of parameters you are estimating. Keep in mind that for even small time series lengths, this N-p number will exceed 30. If you have a t distribution where the number of degrees of freedom exceeds 30, then you can essentially use the N(0,1) distribtuion.

Compare:

1-tcdf(2,30)

ans =

    0.0273

1-normcdf(2,0,1)

ans =

    0.0228

So for your T statistic of 1.85, take your time series length - the number of parameters you are estimating and set that equal to dof, then do:

2*(1-tcdf(1.85,dof))

that gives you the significance level.

[Wayne King]

but again you're probably safe doing

2*(1-normcdf(1.85,0,1))