tsfilter.tsgarch.estimate {tsgarch} | R Documentation |
Model Filtering
Description
Filters new data based on an already estimated model or filters data based on a specification object.
Usage
## S3 method for class 'tsgarch.estimate'
tsfilter(object, y = NULL, newxreg = NULL, newvreg = NULL, ...)
## S3 method for class 'tsgarch.spec'
tsfilter(object, y = NULL, newxreg = NULL, newvreg = NULL, ...)
Arguments
object |
an object of class “tsgarch.estimate” or “tsgarch.spec”. |
y |
an xts vector of new values to filter. Can also be NULL in which case the |
newxreg |
not currently used, |
newvreg |
variance regressors with the same number of rows as y. This can be either a numeric or xts matrix. Only needed if the model was estimated with regressors in the variance equation. |
... |
additional arguments for future expansion. |
Details
The method filters new data and updates the object with this new information so that it can be called recursively as new data arrives. It is also possible to use a specification object with fixed parameters, by appropriately setting the values of the “parmatrix” object in the specification slot. In this case, the returned object will also be of classs “tsgarch.estimate”. If an object of “tsgarch.spec” is used with y not NULL, then the method will first filter the values of the data in the object, generating an object of “tsgarch.estimate” and then call the method again on this new object and the new y values (and optionally any newvreg values). In this way, using either object classes will return the exact same results. The timestamp indices of y must be strictly greater than the maximum timestamp index of the data within the object (i.e. we only filter on new data).
Value
A “tsgarch.estimate” object with updated information.