where and
are the mean and
the precision matrix of the initial state,
is the state
dynamics matrix and
is the precision of the
innovation noise. It is possible that
and/or
are different for each transition instead of being
constant.
The probability distribution is
where
mu (Gaussian-like node or (...,D)-array) – , mean of
,
-dimensional
with plates (…)
Lambda (Wishart-like node or (...,D,D)-array) – , precision matrix of
,
-dimensional with plates (…)
A (Gaussian-like node or (D,D)-array or (...,1,D,D)-array or (...,N-1,D,D)-array) – , state dynamics matrix,
-dimensional with
plates (D,) or (…,1,D) or (…,N-1,D)
nu (gamma-like node or (D,)-array or (...,1,D)-array or (...,N-1,D)-array) – , diagonal elements of the precision of the
innovation process, plates (D,) or (…,1,D) or (…,N-1,D)
n (int, optional) – , the length of the chain. Must be given if
and
are constant over time.
See also
Gaussian
, GaussianARD
, Wishart
, Gamma
, SwitchingGaussianMarkovChain
, VaryingGaussianMarkovChain
, CategoricalMarkovChain
Create GaussianMarkovChain node.
Methods
|
Create GaussianMarkovChain node. |
|
|
|
|
|
Delete this node and the children |
|
Computes gradient with respect to the natural parameters. |
|
|
Return parameters of the VB distribution. |
|
Computes the Riemannian/natural gradient. |
|
|
|
Return True if the node has a plotter |
|
|
|
Set the variable to a random sample from the current distribution. |
|
|
|
|
|
|
Compute the log probability density function Q(X) of this node. |
|
Compute E[ log p(X|parents) - log q(X) ] |
|
|
|
Fix moments, compute f and propagate mask. |
|
Compute the probability density function of this node. |
|
Plot the node distribution using the plotter of the node |
|
Draw a random sample from the distribution. |
|
|
|
|
Set the parameters of the VB distribution. |
|
|
|
|
Print the distribution using standard parameterization. |
|
Attributes
Plate multiplier is applied to messages to parents |