最近在写论文,需要用到GARCH-Jump模型,用EVIEWS的LOGL估计无法得到估计
不知哪位高手可以提供下帮助?最好能给出大概的程序或命令.
非常感谢!


附录是我编的程序,得不到结果

coef(1) mu = .1
coef(1) omega = .1
coef(1) alpha = .1
coef(1) beta = .1
coef(1) la=0.1
coef(1) sita=0
coef(1) u=0
equation eq_temp.ls y c
mu(1) = eq_temp.c(1)
omega(1) = [email=eq_temp.@se^2]eq_temp.@se^2[/email]

' set presample values of expressions in logl
smpl s0
series sig2 = omega(1)
series resma = 0

' set up ARCH likelihood
logl ll1
ll1.append @logl logl
ll1.append m=0
for !j=1 to 1000
ll1.append resma = y-mu(1)-{!j}*u(1)
ll1.append sig2 = omega(1)+alpha(1)*resma(-1)^2 +beta(1)*sig2(-1)+{!j}*(sita(1)^2)
ll1.append [email=z=resma/@sqrt(sig2]z=resma/@sqrt(sig2[/email])
ll1.append [email=l=@dnorm(z)*@dpoisson({!j},la(1))/(sig2^0.5]l=@dnorm(z)*@dpoisson({!j},la(1))/(sig2^0.5[/email])
ll1.append m=l+m
next
ll1.append [email=logl=@log(m]logl=@log(m[/email])

' estimate and display results
smpl s1
ll1.ml(showopts, m=1000, c=1e-5)
show ll1.output