Function kfilter runs the Kalman filter for the given model,
and returns the filtered estimates and one-step-ahead predictions of the
states \(\alpha_t\) given the data up to time \(t\).
kfilter(object, ...)Model object
Ignored.
List containing the log-likelihood (approximate in non-Gaussian case),
one-step-ahead predictions at and filtered
estimates att of states, and the corresponding variances Pt and
Ptt.
For non-Gaussian models, the Kalman filtering is based on the approximate Gaussian model.