State space representation of a time series process, with Kalman filter
Parameters: | k_endog : array_like or integer
k_states : int
k_posdef : int, optional
loglikelihood_burn : int, optional
tolerance : float, optional
results_class : class, optional
**kwargs :
|
---|
Notes
There are several types of options available for controlling the Kalman filter operation. All options are internally held as bitmasks, but can be manipulated by setting class attributes, which act like boolean flags. For more information, see the set_* class method documentation. The options are:
The filter_method and inversion_method options intentionally allow the possibility that multiple methods will be indicated. In the case that multiple methods are selected, the underlying Kalman filter will attempt to select the optional method given the input data.
For example, it may be that INVERT_UNIVARIATE and SOLVE_CHOLESKY are indicated (this is in fact the default case). In this case, if the endogenous vector is 1-dimensional (k_endog = 1), then INVERT_UNIVARIATE is used and inversion reduces to simple division, and if it has a larger dimension, the Cholesky decomposition along with linear solving (rather than explicit matrix inversion) is used. If only SOLVE_CHOLESKY had been set, then the Cholesky decomposition method would always be used, even in the case of 1-dimensional data.
Attributes
design | |
endog | |
obs_cov | |
obs_intercept | |
selection | |
state_cov | |
state_intercept | |
transition |
Methods
bind(endog) | Bind data to the statespace representation |
filter([filter_method, inversion_method, ...]) | Apply the Kalman filter to the statespace model. |
impulse_responses([steps, impulse, ...]) | Impulse response function |
initialize_approximate_diffuse([variance]) | Initialize the statespace model with approximate diffuse values. |
initialize_known(initial_state, ...) | Initialize the statespace model with known distribution for initial state. |
initialize_stationary() | Initialize the statespace model as stationary. |
loglike([loglikelihood_burn]) | Calculate the loglikelihood associated with the statespace model. |
loglikeobs([loglikelihood_burn]) | Calculate the loglikelihood for each observation associated with the statespace model. |
set_conserve_memory([conserve_memory]) | Set the memory conservation method |
set_filter_method([filter_method]) | Set the filtering method |
set_inversion_method([inversion_method]) | Set the inversion method |
set_stability_method([stability_method]) | Set the numerical stability method |
simulate(nsimulations[, measurement_shocks, ...]) | Simulate a new time series following the state space model |
Methods
bind(endog) | Bind data to the statespace representation |
filter([filter_method, inversion_method, ...]) | Apply the Kalman filter to the statespace model. |
impulse_responses([steps, impulse, ...]) | Impulse response function |
initialize_approximate_diffuse([variance]) | Initialize the statespace model with approximate diffuse values. |
initialize_known(initial_state, ...) | Initialize the statespace model with known distribution for initial state. |
initialize_stationary() | Initialize the statespace model as stationary. |
loglike([loglikelihood_burn]) | Calculate the loglikelihood associated with the statespace model. |
loglikeobs([loglikelihood_burn]) | Calculate the loglikelihood for each observation associated with the statespace model. |
set_conserve_memory([conserve_memory]) | Set the memory conservation method |
set_filter_method([filter_method]) | Set the filtering method |
set_inversion_method([inversion_method]) | Set the inversion method |
set_stability_method([stability_method]) | Set the numerical stability method |
simulate(nsimulations[, measurement_shocks, ...]) | Simulate a new time series following the state space model |
Attributes
conserve_memory | int(x=0) -> int or long |
design | |
dtype | (dtype) Datatype of currently active representation matrices |
endog | |
filter_conventional | bool(x) -> bool |
filter_method | int(x=0) -> int or long |
filter_methods | list() -> new empty list |
inversion_method | int(x=0) -> int or long |
inversion_methods | list() -> new empty list |
invert_cholesky | bool(x) -> bool |
invert_lu | bool(x) -> bool |
invert_univariate | bool(x) -> bool |
memory_conserve | bool(x) -> bool |
memory_no_filtered | bool(x) -> bool |
memory_no_forecast | bool(x) -> bool |
memory_no_likelihood | bool(x) -> bool |
memory_no_predicted | bool(x) -> bool |
memory_options | list() -> new empty list |
memory_store_all | bool(x) -> bool |
obs | (array) Observation vector: y~(k\_endog \times nobs) |
obs_cov | |
obs_intercept | |
prefix | (str) BLAS prefix of currently active representation matrices |
selection | |
solve_cholesky | bool(x) -> bool |
solve_lu | bool(x) -> bool |
stability_force_symmetry | bool(x) -> bool |
stability_method | int(x=0) -> int or long |
stability_methods | list() -> new empty list |
state_cov | |
state_intercept | |
time_invariant | (bool) Whether or not currently active representation matrices are |
transition |