LinearGaussianSSM#

cultivars.state_space.linear_gaussian.LinearGaussianSSM[source]#

alias of _LinearGaussianStateSpace

Attributes#

k_endog

Observation dimension p (number of observed series).

k_posdef

State disturbance dimension r.

k_states

State dimension m.

Methods#

filter(y)

Run the Kalman filter over the observation matrix.

loglikelihood(y)

Return the total Gaussian log-likelihood of the data.

mixed_frequency_state_space(coefficients, ...)

Put a latent high-frequency autoregression into observable form.

simulate(n, *[, seed])

Draw a length-n observation path from the model.

simulation_smoother(y, *[, n_sims, seed])

Draw states from p(alpha | y) via the Durbin-Koopman (2002) smoother.

smooth(y)

Run the Durbin-Koopman state smoother.

stationary_covariance(transition, selection, ...)

Solve the discrete Lyapunov equation for the stationary state covariance.