Forecast GARCH (variance) series

[student07]

Hello,

I am trying to forecast the GARCH series from a GARCH equation.
The object reference writes the following about this:

eq_name.fit(options) yhat [y_se]
eq_name.fit(options) yhat [y_se y_var]
Following the fit keyword, you should type a name for the forecast series and, optionally, a
name for the series containing the standard errors. For ARCH specifications, you may use the second form of the command, and optionally include a name for the conditional variance
series.

Am I correct in concluding that if I write for instance

Code: Select all

test.fit yhat a b

that then a contains standard errors of the FORECAST for the mean equation (so not of the GARCH series) and b contains the forecasted GARCH series (so the forecasted VARIANCE from the GARCH equation)? Also I assume that the last argument is really the forecasted variance and not the square root of that (i.e. the volatility)?

Lastly, I noticed that some times I get negative values out of the last argument (what I assumed to be the forecasted GARCH variance) so maybe I am wrong about its meaning.

Thanks in advance.

[trubador]

You are correct. But resulting garch series should not include negative values. In such cases, you may want to check the estimated model and make sure that diagnostics are fine.

[student07]

You are correct. But resulting garch series should not include negative values. In such cases, you may want to check the estimated model and make sure that diagnostics are fine.

Thank you. What kind of diagnostics should I be looking at (p-values seem fine)? (FYI, I have no predictors in my specification, it is a simple GARCH(1,1) model).