ARMA-GARCH estimation with EGB2 distribution

Hi,

I want to estimate an ARMA-GARCH model by using the EGB2 distribution. The model I want to estimate is:

The loglikelihood based on the EGB2 distribution is:

Here .

I minimize the following function:

    loglike = 0;
    for i = 1:length(R)-8
        loglike = loglike + (p*sqrt(OMEGA)*eps(i)/sqrt(sigma_2(i)) - 0.5*log(sigma_2(i)) - (p+q)*log(1+exp(sqrt(OMEGA)*eps(i)/sqrt(sigma_2(i))) + DELTA));
    end 
    
    logL = -(length(eps)*(log(sqrt(OMEGA))-log(betaFunc(p,q))+p*DELTA) + loglike);

The constraints are that and . The outcome must be a value of the loglikelihood around -3000. But I get 2.5+e07.

What is going wrong? Must I have more restrictions?

Thanks in advance!