Generate state series or state standard error series from an estimated sspace object.
Options allow you to generate one-step ahead, filtered, or smoothed values for the states and the state standard errors.
Syntax
sspace_name.makestates(options) [name_spec]
Follow the object name with a period and the makestate keyword, options to determine the output type, and a list of names or a wildcard expression identifying the series to hold the output. If a list is used to identify the targets, the number of target series must match the number of names implied by the keyword.
Options
t=output_type (default=“pred”) | Defines output type: one-step ahead state predictions (“pred”), RMSE of the one-step ahead state predictions (“predse”), error in one-step ahead state predictions (“resid”), RMSE of the one-step ahead state prediction (“residse”), filtered states (“filt”), RMSE of the filtered states (“filtse”), standardized one-step ahead prediction residual (“stdresid”), smoothed states (“smooth”), RMSE of the smoothed states (“smoothse”), estimate of the disturbances (“disturb”), RMSE of the estimate of the disturbances (“disturbse”), standardized estimate of the disturbances (“stddisturb”). |
n=group_name | Name of group to hold newly created series. |
prompt | Force the dialog to appear from within a program. |
Examples
ss1.makestates(t=smooth) sm*
produces smoothed states in the series with names beginning with “sm”, and ending with the name of the state dependent variable.
ss2.makestates(t=pred, n=pred_states) sig1 sig2 sig3
creates a group named PRED_STATES which contains the one-step ahead state predictions in series SIG1, SIG2, and SIG3.
Cross-references
See
“State Space Models and the Kalman Filter” for details on state space models. For additional discussion of wildcards, see
Appendix A. “Wildcards”.