Search found 3 matches
- Sun Sep 07, 2014 3:38 am
- Forum: Estimation
- Topic: about garch model's first observation
- Replies: 0
- Views: 2446
about garch model's first observation
Hello, and after reading some paper and code, then try, I find it is easily to show "missing value in the first obsevation" I think it is because the first observation has not former perid value to calculate this perid's var and res, is it right? and after some reading I think there is 2 w...
- Sun Sep 07, 2014 2:45 am
- Forum: Estimation
- Topic: why there are two var?
- Replies: 2
- Views: 8509
Re: why there are two var?
Impossible to say without seeing the rest of the program. formula is ht=ω+αut-1^2+δ1*G1*(ht-1)+β*ht-1+λ0d0 load return series d1=0 smpl @first @first d1=1 smpl @all coef(1) mu(1)=0.0000309 coef(1) omega(1)=0.0075 coef(1) alpha(1)=0.05 coef(1) gamma(1)=0.00755 coef(1) du(1)=0.0123 coef(1) delta(1)=0...
- Sat Sep 06, 2014 2:07 am
- Forum: Estimation
- Topic: why there are two var?
- Replies: 2
- Views: 8509
why there are two var?
I found a paper’s code about stgarch which formula is ht=ω+αut-1^2+δ1*G1*(ht-1)+β*ht-1+λ0d0 and the writer writes the code like this .... var=@recode(d1=1,omega(2)/(1-alpha(2)),omega(2)+delata*@gammainc(var(-1),gamma+alpha(2)*res(-1)^2) var=@recode(d1=1,0.05,du(1)*dummy+omga(2)+delta(1)*@gammainc(va...
