Calculate the marginal likelihood of phi given the values of the latent-state and other static parameters
Calculate the marginal likelihood of phi given the values of the latent-state and other static parameters
a sample of the latent state of an AR(1) DLM
the static parameters of a DLM
autoregressive
A single step of a Gibbs Sampler
A single step of a Gibbs Sampler
the model containing the definition of the observation matrix F_t and system evolution matrix G_t
the prior distribution on the observation noise matrix, V
the prior distribution on the system noise matrix, W
an array of Data containing the observed time series
Use metropolis hastings to determine the initial state distribution x0 ~ N(m0, C0)
Use metropolis hastings to determine the initial state distribution x0 ~ N(m0, C0)
a proposal distribution for the parameters of the initial state
a DLM model specification
the prior distribution of the observation noise matrix
the prior distribution of the system noise matrix
a vector of observations
Calculate the marginal likelihood for metropolis hastings
A metropolis step for a DLM
A metropolis step for a DLM
a DLM model
the currently sampled state of the DLM
a symmetric proposal distribution for the parameters of a DLM
a function from Parameters => Rand[Parameters] which performs a metropolis step to be used in a Markov Chain
Return a Markov chain using Gibbs Sampling to determine the values of the system and observation noise covariance matrices, W and V
Return a Markov chain using Gibbs Sampling to determine the values of the system and observation noise covariance matrices, W and V
the model containing the definition of the observation matrix F_t and system evolution matrix G_t
the prior distribution on the observation noise matrix, V
the prior distribution on the system noise matrix, W
the initial parameters of the Markov Chain
an array of Data containing the observed time series
a Process
Sample the (diagonal) observation noise covariance matrix from an Inverse Gamma distribution
Sample the (diagonal) observation noise covariance matrix from an Inverse Gamma distribution
an Inverse Gamma prior distribution for each variance element of the observation matrix
the observed values of the time series
a sample of the DLM state
the posterior distribution over the diagonal observation matrix
Sample the autoregressive parameter with a Beta Prior and proposal distribution
Sample the diagonal system matrix for an irregularly observed DLM
Update an autoregressive model with a new value of the autoregressive parameter