State for the Kalman Filter
State for the Kalman Filter
the current timestep
the posterior mean of the latent state
the posterior covariance of the latent state
the prior mean of the latent state
the prior covariance of the latent state
the one step predicted observation mean, not present at the first timestep
the one step predicted observation covariance, not present at the first timestep
the current log-likelihood of all the observations up until the current time
Advance the state mean and variance to the a-priori value of the state at time t
Advance the state mean and variance to the a-priori value of the state at time t
the system matrix, a function from a time increment to DenseMatrix
the a-posteriori mean of the state at time t-1
the a-posteriori covariance of the state at time t-1
the time increment
the system noise matrix
the a-priori mean and covariance of the state at time t
Calculate the conditional likelihood of the
Run the Kalman Filter over an array of data
Remove optional data from an observation vector
Remove optional data from an observation vector
a vector containing optional observations
a vector containing only the observations which are there
Get the index of the non-missing data
Get the index of the non-missing data
a vector of observations possibly containing missing data
a vector containing the indices of non-missing observations
Calculate the marginal likelihood of a DLM using a kalman filter
Build observation matrix for potentially missing data
Build observation error variance matrix for potentially missing data
Perform a one-step prediction taking into account missing data in the observations, this alters the size of the F-matrix
Perform a one-step prediction taking into account missing data in the observations, this alters the size of the F-matrix
the observation matrix
the a-priori mean state at time t
the a-priori covariance of the state at time t
the current time
the observation variance
the observation at time t
Perform a one-step prediction
Step the Kalman Filter a single Step
Update the state using Joseph Form Update given the newly observed data
Update the state using Joseph Form Update given the newly observed data
the observation matrix
the a priori state mean at time t
the a priori state variance at time t
the actual observation at time t
the variance of the measurement noise
the current log-likelihood
the posterior mean and variance of the latent state at time t