 |
Download a pdf copy of this help file here |
|
Download a pdf copy of this help file here |
See also: Time series in ModelRisk, Vose Multivariate Time Series window
VoseTimeMultiAR1({a},{b},{c}, LogReturn,{LastValues},Data_in_rows)
Array function that simulates from a Multivariate GARCH model in the BEKK parametrisation. There are many generalizations of GARCH models to higher dimensions, each simplifying the most general possible approach in a number of ways to keep the parameter count under control. Among these, the BEKK parameterization is generally considered the most sensible one, though it has the drawback that its parameter matrices lack an obvious interpretation.
{a} - Autoregressive parameters matrix.
{b} - Moving-Average parameters matrix.
{c} - covariance matrix.
LogReturn - Optional parameter specifying whether to return the actual time series (FALSE, default) or the log returns.
{LastValues} - array of last known historic values to start forecast from.
Data_in_rows - optional parameter that specifies if the data is in rows (TRUE) or columns (FALSE, default).
VoseTimeMultiBEKK - generates an array of random values from this time series.
VoseTimeMultiBEKKObject - creates an Object for this time series.
VoseTimeMultiBEKKFit - generates an array with random values from this time series fitted to data.
VoseTimeMultiBEKKFitObject - creates an Object for this time series fitted to data.