This function implements the full information propagation step under mixture Kalman filter with full information proposal distribution and Rao-Blackwellization, no delay.
MKF.Full.RB(
MKFstep.Full.RB,
nobs,
yy,
mm,
par,
II.init,
mu.init,
SS.init,
xdim,
ydim,
resample.sch
)The function returns a list with components:
the fitted value.
the fitted value using Rao-Blackwellization.
the estimated indicators.
the estimated indicators using Rao-Blackwellization.
a function that performs one step propagation under mixture Kalman filter, with full information proposal distribution.
Its input includes (mm,II,mu,SS,logww,yyy,par,xdim,ydim), where
II, mu, and SS are the indicators and its corresponding mean and variance matrix of the Kalman filter components in the last iterations.
logww is the log weight of the last iteration. yyy is the
observation at current time step. It should return the Rao-Blackwellization estimation of the mean and variance.
the number of observations T.
the observations with T columns and ydim rows.
the Monte Carlo sample size m.
a list of parameter values to pass to Sstep.
the initial indicators.
the initial mean.
the initial variance.
the dimension of the state varible x_t.
the dimension of the observation y_t.
a binary vector of length nobs, reflecting the resampling schedule. resample.sch[i]= 1 indicating resample should be carried out at step i.
Tsay, R. and Chen, R. (2018). Nonlinear Time Series Analysis. John Wiley & Sons, New Jersey.