http://www.mathworks.com/access/helpdesk/help/toolbox/garch/forecasting10.html#81184
 
this data also in gretl
 
you should do it
open b-g
garch 1 1; Y
 
and i don't know how do i forecast..
 
forecasting future volatility...with mathlab.
I am having trouble interpreting the output of garchpred,for example,
 
spec = garchset('P',1,'Q',1,'TolCon',1e-6);
garchfit(spec,bg);
[coeff,errors,LLF,innovations,sigmas] = garchfit(bg);
[sigmaForecast,meanForecast]  = garchpred(coeff,bg,10);

[sigmaForecast,meanForecast]

VL = coeff.K/(1-coeff.ARCH-coeff.GARCH)

newvar = sqrt(VL + (coeff.ARCH+coeff.GARCH)^[1 2 3 4 5 6 7 8 9 10]'*(sigmas(1974)^2-VL))
 
% this equation is refrenced Options, Future and other Derivatives 6th page 472 equation 19.13(JOHN C. HULL)
 
sigmaForecast =

    0.3834
    0.3895
    0.3953
    0.4008
    0.4060
    0.4110
    0.4156
    0.4200
    0.4242
    0.4282

>> newvar = sqrt(VL + (coeff.ARCH+coeff.GARCH ).^[1 2 3 4 5 6 7 8 9 10]'*(sigmas(1974)^2-VL))

newvar =

    0.3477
    0.3559
    0.3637
    0.3710
    0.3778
    0.3843
    0.3904
    0.3961
    0.4016
    0.4067

>>
 
why two result is different..  and during 10days volatility is mean(sigmaForecast ). is correct?
 
i'm still in garch world -_-; anybody help me...