GARCH Toolbox

Forecasting

This section uses the estimated default model and the XYZ Corporation, from the section Simulation, to demonstrate the use of the forecasting function garchpred.

garchpred computes minimum-mean-square-error (MMSE) forecasts of the conditional mean and conditional standard deviation of the returns {y_t} in each period over a user-specified forecast horizon.

Specifically, this section discusses:

• Computing a Forecast
• Computing Root Mean Square Errors (RMSE)
• Asymptotic Behavior for Long-Range Forecast Horizons

Note    Example results in this section are printed in Short E numeric format for readability. Select File > Preferences... > General > Short E before starting the example to duplicate these printed results.

A General Simulation Example

Computing a Forecast