problem with egarch 2,2

[karl5914]

I am trying to implement my EGARCH 2,2 model with two components.
and I entered the follwoing commands:
nonlin A11 A22 A33 A44 A2 A3 A4

SET Q / = %seesq**0.5
SET H / = EXP(RANGE - 0.43)
SET U / = 0.0
compute A11=1, A22=0.9,A33=0.05,A44=-0.1, A2=0.12,A3=0.01,A4=-0.1
frml res = (RANGE - 0.43 - LOG(q))/0.29
frml qf = exp(A11+A22*log(q{1}) + A33*u{1} + A44*E{1}/q{1})
frml HF = exp(qf{1}+A2*log(h{1})+A3*u{1} + A4*E{1}/h{1})
frml logl = h(t)=hf(t),(Q(t)=QF(t)),(u(t)=res(t)),-0.5*(log(0.29**2) + u(t)**2/.29**2)
maximize(method=GENETIC,iters=100) LOGL 3 500
maximize(method=bfgs, iters=100) LOGL 3 500

but my result does not have the t-test for the last three variables like:

MAXIMIZE - Estimation by BFGS
Convergence in 8 Iterations. Final criterion was 0.0000000 <= 0.0000100
Daily(5) Data From 1987:01:05 To 1988:11:30
Usable Observations 498
Function Value 616.40462187

Variable Coeff Std Error T-Stat Signif
*******************************************************************************
1. A11 -0.428450517 0.000069382 -6175.24257 0.00000000
2. A22 0.000215242 0.008749363 0.02460 0.98037336
3. A33 0.104713782 0.209460362 0.49992 0.61713021
4. A44 0.001762898 0.032202527 0.05474 0.95634234
5. A2 0.328352033 0.000000000 0.00000 0.00000000
6. A3 0.118726418 0.000000000 0.00000 0.00000000
7. A4 -1.335753740 0.000000000 0.00000 0.00000000

Would anyone know what has went wrong?

[TomDoan]

I am trying to implement my EGARCH 2,2 model with two components.
and I entered the follwoing commands:

Code: Select all

nonlin A11 A22 A33 A44 A2 A3 A4

SET Q / = %seesq**0.5
SET H / = EXP(RANGE - 0.43)
SET U / = 0.0
compute A11=1, A22=0.9,A33=0.05,A44=-0.1, A2=0.12,A3=0.01,A4=-0.1
frml res = (RANGE - 0.43 - LOG(q))/0.29
frml qf = exp(A11+A22*log(q{1}) + A33*u{1} + A44*E{1}/q{1})
frml HF = exp(qf{1}+A2*log(h{1})+A3*u{1} + A4*E{1}/h{1})
frml logl = h(t)=hf(t),(Q(t)=QF(t)),(u(t)=res(t)),-0.5*(log(0.29**2) + u(t)**2/.29**2)
maximize(method=GENETIC,iters=100) LOGL 3 500
maximize(method=bfgs, iters=100) LOGL 3 500

but my result does not have the t-test for the last three variables like:

MAXIMIZE - Estimation by BFGS
Convergence in 8 Iterations. Final criterion was 0.0000000 <= 0.0000100
Daily(5) Data From 1987:01:05 To 1988:11:30
Usable Observations 498
Function Value 616.40462187

Variable Coeff Std Error T-Stat Signif
*******************************************************************************
1. A11 -0.428450517 0.000069382 -6175.24257 0.00000000
2. A22 0.000215242 0.008749363 0.02460 0.98037336
3. A33 0.104713782 0.209460362 0.49992 0.61713021
4. A44 0.001762898 0.032202527 0.05474 0.95634234
5. A2 0.328352033 0.000000000 0.00000 0.00000000
6. A3 0.118726418 0.000000000 0.00000 0.00000000
7. A4 -1.335753740 0.000000000 0.00000 0.00000000

Would anyone know what has went wrong?

Where does H enter your likelihood? It looks as if you're hard coding a variance of .29 rather than using H.

[karl5914]

the dependent variable is distrbuted as n( 0.43+lnh, 0.29**2)

[TomDoan]

the dependent variable is distrbuted as n( 0.43+lnh, 0.29**2)

Code: Select all

frml res = (RANGE - 0.43 - LOG(q))/0.29

You have log(q), not log(h) in this. You also don't want the /0.29 here, since you're dividing by that again in the log likelihood calculation.

[karl5914]

Thanks it works