Dynamic Linear Model Action Set
Provides an action for fitting dynamic linear models
dynamicLinear Action
Learns and infers dynamic linear models.
Summary: Input and Output Tables
If a row includes a subparameter, you can specify the name, caslib, and so on in the subparameter. Otherwise, you can specify the name, caslib, and so on in the parameter.
|
Parameter |
Subparameter |
Description |
|---|---|---|
|
required parameterdata |
— |
specifies the input data table. |
|
inDiscountFactor |
specifies the model's discount factors. | |
|
inMean, inCovariance (and nested parameter diagonal), (and nested parameter full), inGamma |
specifies the parameters for the initial distribution. | |
|
inAdjacency |
specifies the model's parental sets. |
|
Parameter |
Subparameter |
Description |
|---|---|---|
|
outVerification |
specifies the model's discount factors. | |
|
out, outSim, outCovariance, outFilterInfo (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma), (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma), (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma), (and nested parameter klDivergence), (and nested parameter theta), (and nested parameter nu), (and nested parameter lambda), (and nested parameter theta), (and nested parameter lambda) |
specifies the options that are related to the filtering problem, which aims to find the posterior distribution of the current state, based on observations that precede and include the current observation. | |
|
out, outSim, outCovariance |
specifies the options that are related to the forecasting problem, which aims to determine the probability distribution of future states. | |
|
outVerification (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma) |
specifies the parameters for the initial distribution. | |
|
outVerification |
specifies the model's parental sets. |
Parameter Descriptions
* data={castable}
specifies the input data table.
For more information about specifying the data parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | table |
|---|
discountFactor={dlmDfFactorStmt}
specifies the model's discount factors.
The dlmDfFactorStmt value can be one or more of the following:
beta={betaStmt}
specifies the model's stochastic volatility discount factors.
The betaStmt value can be one or more of the following:
defaultValue=double
specifies the default beta parameter value for all variables.
| Alias | def |
|---|---|
| Default | 0.95 |
| Range | (0, 1) |
positions={integer-1 <, integer-2, ...>}
specifies the positions of variables for assigning beta parameter values different from the default.
| Alias | pos |
|---|
values={double-1 <, double-2, ...>}
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
deltaGamma={deltaGammaStmt}
specifies the model's discount factors for regressors in the parental sets.
The deltaGammaStmt value can be one or more of the following:
defaultValue=double
specifies the default deltaGamma parameter value for variables in the parental sets.
| Alias | def |
|---|---|
| Default | 0.995 |
| Range | (0, 1) |
positions={integer-1 <, integer-2, ...>}
specifies the positions of variables for assigning deltaGamma parameter values different from the default.
| Alias | pos |
|---|
values={double-1 <, double-2, ...>}
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
deltaPhi={deltaPhiStmt}
specifies the model's discount factors for regressors not in the parental sets.
The deltaPhiStmt value can be one or more of the following:
defaultValue=double
specifies the default deltaPhi parameter value for regressors not in the parental sets.
| Alias | def |
|---|---|
| Default | 0.995 |
| Range | (0, 1) |
positions={integer-1 <, integer-2, ...>}
specifies the positions of variables for assigning deltaPhi parameter values different from the default.
| Alias | pos |
|---|
values={double-1 <, double-2, ...>}
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
inDiscountFactor={castable}
specifies the initial values for all discount factors.
For more information about specifying the inDiscountFactor parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | inDF |
|---|
outVerification={casouttable}
specifies the table to use for validating the input discount factors.
For more information about specifying the outVerification parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outVerify |
|---|
filter={dlmFilterStmt}
specifies the options that are related to the filtering problem, which aims to find the posterior distribution of the current state, based on observations that precede and include the current observation.
The dlmFilterStmt value can be one or more of the following:
back=integer
specifies the number of observations before the end of the data at which the forecasts begin.
| Default | 1 |
|---|---|
| Minimum value | 0 |
lead=integer
specifies the number of the multistep forecast values to compute during filtering.
| Default | 1 |
|---|---|
| Minimum value | 1 |
out={casouttable}
writes in-sample forecasting results and statistics to an output table.
For more information about specifying the out parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
outCovariance={casouttable}
writes the covariances of samples of in-sample forecasting observations to an output table.
For more information about specifying the outCovariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outCov |
|---|
outFilterInfo={outFilterInfoStmt}
specifies which filtering information to write to output tables.
The outFilterInfoStmt value can be one or more of the following:
klDivergence={casouttable}
specifies the output table for Kullback-Leibler divergence information in the decoupling step of updating the distribution from naive posterior to posterior.
For more information about specifying the klDivergence parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | klDiv |
|---|
naivePosterior={filterNaivePostStmt}
specifies which parameters of the naive posterior distribution to output.
| Alias | naivePost |
|---|
The filterNaivePostStmt value can be one or more of the following:
covariance={casouttable}
specifies the output table for the covariances of naive posterior normal-gamma distributions.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
gamma={casouttable}
specifies the output table for the shape and rate parameters of naive posterior normal-gamma distributions.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
mean={casouttable}
specifies the output table for the means of naive posterior normal-gamma distributions.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
posterior={filterPostStmt}
specifies which parameters of the posterior distribution to output.
| Alias | post |
|---|
The filterPostStmt value can be one or more of the following:
covariance={casouttable}
specifies the output table for the covariances of posterior normal-gamma distributions.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
gamma={casouttable}
specifies the output table for the shape and rate parameters of posterior normal-gamma distributions.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
mean={casouttable}
specifies the output table for the means of posterior normal-gamma distributions.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
predictionSample={filterPredRawSmplStmt}
specifies which raw random samples to output during forecasting.
| Alias | predSmpl |
|---|
The filterPredRawSmplStmt value can be one or more of the following:
lambda={casouttable}
specifies the output table for raw random samples of precisions during forecasting.
For more information about specifying the lambda parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
nu={casouttable}
specifies the output table for raw random samples of observation errors during forecasting.
For more information about specifying the nu parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
theta={casouttable}
specifies the output table for raw random samples of state vectors during forecasting.
For more information about specifying the theta parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
prior={filterPriorStmt}
specifies which parameters of the prior distribution to output.
The filterPriorStmt value can be one or more of the following:
covariance={casouttable}
specifies the output table for the covariances of prior normal-gamma distributions.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
gamma={casouttable}
specifies the output table for the shape and rate parameters of prior normal-gamma distributions.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
mean={casouttable}
specifies the output table for the means of prior normal-gamma distributions.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
recouplingSample={filterImpSmplRawSmplStmt}
specifies which raw random samples to output during the recoupling of importance sampling.
| Alias | recSmpl |
|---|
The filterImpSmplRawSmplStmt value can be one or more of the following:
lambda={casouttable}
specifies the output table for raw random samples of precisions during the recoupling of importance sampling.
For more information about specifying the lambda parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
theta={casouttable}
specifies the output table for raw random samples of state vectors during the recoupling of importance sampling.
For more information about specifying the theta parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
outSim={casouttable}
writes samples of in-sample forecasting observations to an output table.
For more information about specifying the outSim parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
percent={double-1 <, double-2, ...>}
specifies percentile points for the summary of filtering results.
| Alias | pct |
|---|
forecast={dlmForecastStmt}
specifies the options that are related to the forecasting problem, which aims to determine the probability distribution of future states.
The dlmForecastStmt value can be one or more of the following:
lead=integer
specifies the number of the multistep forecast values to compute during forecasting.
| Default | 1 |
|---|---|
| Minimum value | 1 |
out={casouttable}
writes out-of-sample forecasting results and statistics to an output table.
For more information about specifying the out parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
outCovariance={casouttable}
writes the covariances of samples of out-of-sample forecasting observations to an output table.
For more information about specifying the outCovariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outCov |
|---|
outSim={casouttable}
writes samples of out-of-sample forecasting observations to an output table.
For more information about specifying the outSim parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
percent={double-1 <, double-2, ...>}
specifies percentile points for the summary of forecasting results.
| Alias | pct |
|---|
* id="variable-name"
identifies observations in the input data table by specifying a variable for the time series data.
initialDistribution={dlmInitDistStmt}
specifies the parameters for the initial distribution.
The dlmInitDistStmt value can be one or more of the following:
inCovariance={covStmt}
specifies the starting covariance settings for the initial distribution.
| Alias | inCov |
|---|
The covStmt value can be one or more of the following:
diagonal={castable}
specifies the initial variance values (diagonal) for the initial distribution's covariances.
For more information about specifying the diagonal parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | diag |
|---|
full={castable}
specifies the initial covariance values by using a row-oriented table.
For more information about specifying the full parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
inGamma={castable}
specifies the initial values for the shape and rate parameters of the normal-gamma distribution.
For more information about specifying the inGamma parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
inMean={castable}
specifies the initial values of means by using a column-oriented table.
For more information about specifying the inMean parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
mean=double
specifies the default initial value of all means.
| Default | 0 |
|---|
outVerification={checkInitialDistStmt}
outputs the initial distribution to tables for validation.
| Alias | outVerify |
|---|
The checkInitialDistStmt value can be one or more of the following:
covariance={casouttable}
specifies the output table for the covariances of the initial distribution.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
gamma={casouttable}
specifies the output table for the shape and rate parameters of the initial distribution.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
mean={casouttable}
specifies the output table for the means of the initial distribution.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
rate={rateStmt}
specifies the starting setting for the initial distribution's rate parameter.
The rateStmt value can be one or more of the following:
defaultValue=double
specifies the default rate parameter value for all variables.
| Alias | def |
|---|---|
| Default | 5 |
| Minimum value (exclusive) | 0 |
positions={integer-1 <, integer-2, ...>}
specifies the positions of variables for assigning rate parameter values different from the default.
| Alias | pos |
|---|
values={double-1 <, double-2, ...>}
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
shape={shapeStmt}
specifies the starting setting for the initial distribution's shape parameter.
The shapeStmt value can be one or more of the following:
defaultValue=double
specifies the default shape parameter value for all variables.
| Alias | def |
|---|---|
| Default | 10 |
| Minimum value (exclusive) | 1 |
positions={integer-1 <, integer-2, ...>}
specifies the positions of variables for assigning shape parameter values different from the default.
| Alias | pos |
|---|
values={double-1 <, double-2, ...>}
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
model={dlmModelStmt}
specifies the options that are related to the model specification.
* variables={"variable-name-1" <, "variable-name-2", ...>}
specifies the variables to use for the model.
| Alias | vars |
|---|
nSimulations=integer
specifies the number of simulations in filtering and forecasting.
| Alias | nSim |
|---|---|
| Default | 10000 |
| Minimum value | 1 |
parentalSet={dlmParSetStmt}
specifies the model's parental sets.
The dlmParSetStmt value can be one or more of the following:
inAdjacency={castable}
specifies the adjacency matrix of the parental sets.
For more information about specifying the inAdjacency parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | inAdj |
|---|
outVerification={casouttable}
specifies the output table to use for verifying the input adjacency matrices.
For more information about specifying the outVerification parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outVerify |
|---|
seed=64-bit-integer
specifies a nonnegative integer to use as the seed for generating random number sequences.
| Default | 0 |
|---|---|
| Minimum value | 0 |
dynamicLinear Action
Learns and infers dynamic linear models.
Summary: Input and Output Tables
If a row includes a subparameter, you can specify the name, caslib, and so on in the subparameter. Otherwise, you can specify the name, caslib, and so on in the parameter.
|
Parameter |
Subparameter |
Description |
|---|---|---|
|
required parameterdata |
— |
specifies the input data table. |
|
inDiscountFactor |
specifies the model's discount factors. | |
|
inMean, inCovariance (and nested parameter diagonal), (and nested parameter full), inGamma |
specifies the parameters for the initial distribution. | |
|
inAdjacency |
specifies the model's parental sets. |
|
Parameter |
Subparameter |
Description |
|---|---|---|
|
outVerification |
specifies the model's discount factors. | |
|
out, outSim, outCovariance, outFilterInfo (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma), (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma), (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma), (and nested parameter klDivergence), (and nested parameter theta), (and nested parameter nu), (and nested parameter lambda), (and nested parameter theta), (and nested parameter lambda) |
specifies the options that are related to the filtering problem, which aims to find the posterior distribution of the current state, based on observations that precede and include the current observation. | |
|
out, outSim, outCovariance |
specifies the options that are related to the forecasting problem, which aims to determine the probability distribution of future states. | |
|
outVerification (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma) |
specifies the parameters for the initial distribution. | |
|
outVerification |
specifies the model's parental sets. |
Parameter Descriptions
* data={castable}
specifies the input data table.
For more information about specifying the data parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | table |
|---|
discountFactor={dlmDfFactorStmt}
specifies the model's discount factors.
The dlmDfFactorStmt value can be one or more of the following:
beta={betaStmt}
specifies the model's stochastic volatility discount factors.
The betaStmt value can be one or more of the following:
defaultValue=double
specifies the default beta parameter value for all variables.
| Alias | def |
|---|---|
| Default | 0.95 |
| Range | (0, 1) |
positions={integer-1 <, integer-2, ...>}
specifies the positions of variables for assigning beta parameter values different from the default.
| Alias | pos |
|---|
values={double-1 <, double-2, ...>}
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
deltaGamma={deltaGammaStmt}
specifies the model's discount factors for regressors in the parental sets.
The deltaGammaStmt value can be one or more of the following:
defaultValue=double
specifies the default deltaGamma parameter value for variables in the parental sets.
| Alias | def |
|---|---|
| Default | 0.995 |
| Range | (0, 1) |
positions={integer-1 <, integer-2, ...>}
specifies the positions of variables for assigning deltaGamma parameter values different from the default.
| Alias | pos |
|---|
values={double-1 <, double-2, ...>}
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
deltaPhi={deltaPhiStmt}
specifies the model's discount factors for regressors not in the parental sets.
The deltaPhiStmt value can be one or more of the following:
defaultValue=double
specifies the default deltaPhi parameter value for regressors not in the parental sets.
| Alias | def |
|---|---|
| Default | 0.995 |
| Range | (0, 1) |
positions={integer-1 <, integer-2, ...>}
specifies the positions of variables for assigning deltaPhi parameter values different from the default.
| Alias | pos |
|---|
values={double-1 <, double-2, ...>}
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
inDiscountFactor={castable}
specifies the initial values for all discount factors.
For more information about specifying the inDiscountFactor parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | inDF |
|---|
outVerification={casouttable}
specifies the table to use for validating the input discount factors.
For more information about specifying the outVerification parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outVerify |
|---|
filter={dlmFilterStmt}
specifies the options that are related to the filtering problem, which aims to find the posterior distribution of the current state, based on observations that precede and include the current observation.
The dlmFilterStmt value can be one or more of the following:
back=integer
specifies the number of observations before the end of the data at which the forecasts begin.
| Default | 1 |
|---|---|
| Minimum value | 0 |
lead=integer
specifies the number of the multistep forecast values to compute during filtering.
| Default | 1 |
|---|---|
| Minimum value | 1 |
out={casouttable}
writes in-sample forecasting results and statistics to an output table.
For more information about specifying the out parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
outCovariance={casouttable}
writes the covariances of samples of in-sample forecasting observations to an output table.
For more information about specifying the outCovariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outCov |
|---|
outFilterInfo={outFilterInfoStmt}
specifies which filtering information to write to output tables.
The outFilterInfoStmt value can be one or more of the following:
klDivergence={casouttable}
specifies the output table for Kullback-Leibler divergence information in the decoupling step of updating the distribution from naive posterior to posterior.
For more information about specifying the klDivergence parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | klDiv |
|---|
naivePosterior={filterNaivePostStmt}
specifies which parameters of the naive posterior distribution to output.
| Alias | naivePost |
|---|
The filterNaivePostStmt value can be one or more of the following:
covariance={casouttable}
specifies the output table for the covariances of naive posterior normal-gamma distributions.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
gamma={casouttable}
specifies the output table for the shape and rate parameters of naive posterior normal-gamma distributions.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
mean={casouttable}
specifies the output table for the means of naive posterior normal-gamma distributions.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
posterior={filterPostStmt}
specifies which parameters of the posterior distribution to output.
| Alias | post |
|---|
The filterPostStmt value can be one or more of the following:
covariance={casouttable}
specifies the output table for the covariances of posterior normal-gamma distributions.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
gamma={casouttable}
specifies the output table for the shape and rate parameters of posterior normal-gamma distributions.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
mean={casouttable}
specifies the output table for the means of posterior normal-gamma distributions.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
predictionSample={filterPredRawSmplStmt}
specifies which raw random samples to output during forecasting.
| Alias | predSmpl |
|---|
The filterPredRawSmplStmt value can be one or more of the following:
lambda={casouttable}
specifies the output table for raw random samples of precisions during forecasting.
For more information about specifying the lambda parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
nu={casouttable}
specifies the output table for raw random samples of observation errors during forecasting.
For more information about specifying the nu parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
theta={casouttable}
specifies the output table for raw random samples of state vectors during forecasting.
For more information about specifying the theta parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
prior={filterPriorStmt}
specifies which parameters of the prior distribution to output.
The filterPriorStmt value can be one or more of the following:
covariance={casouttable}
specifies the output table for the covariances of prior normal-gamma distributions.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
gamma={casouttable}
specifies the output table for the shape and rate parameters of prior normal-gamma distributions.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
mean={casouttable}
specifies the output table for the means of prior normal-gamma distributions.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
recouplingSample={filterImpSmplRawSmplStmt}
specifies which raw random samples to output during the recoupling of importance sampling.
| Alias | recSmpl |
|---|
The filterImpSmplRawSmplStmt value can be one or more of the following:
lambda={casouttable}
specifies the output table for raw random samples of precisions during the recoupling of importance sampling.
For more information about specifying the lambda parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
theta={casouttable}
specifies the output table for raw random samples of state vectors during the recoupling of importance sampling.
For more information about specifying the theta parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
outSim={casouttable}
writes samples of in-sample forecasting observations to an output table.
For more information about specifying the outSim parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
percent={double-1 <, double-2, ...>}
specifies percentile points for the summary of filtering results.
| Alias | pct |
|---|
forecast={dlmForecastStmt}
specifies the options that are related to the forecasting problem, which aims to determine the probability distribution of future states.
The dlmForecastStmt value can be one or more of the following:
lead=integer
specifies the number of the multistep forecast values to compute during forecasting.
| Default | 1 |
|---|---|
| Minimum value | 1 |
out={casouttable}
writes out-of-sample forecasting results and statistics to an output table.
For more information about specifying the out parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
outCovariance={casouttable}
writes the covariances of samples of out-of-sample forecasting observations to an output table.
For more information about specifying the outCovariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outCov |
|---|
outSim={casouttable}
writes samples of out-of-sample forecasting observations to an output table.
For more information about specifying the outSim parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
percent={double-1 <, double-2, ...>}
specifies percentile points for the summary of forecasting results.
| Alias | pct |
|---|
* id="variable-name"
identifies observations in the input data table by specifying a variable for the time series data.
initialDistribution={dlmInitDistStmt}
specifies the parameters for the initial distribution.
The dlmInitDistStmt value can be one or more of the following:
inCovariance={covStmt}
specifies the starting covariance settings for the initial distribution.
| Alias | inCov |
|---|
The covStmt value can be one or more of the following:
diagonal={castable}
specifies the initial variance values (diagonal) for the initial distribution's covariances.
For more information about specifying the diagonal parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | diag |
|---|
full={castable}
specifies the initial covariance values by using a row-oriented table.
For more information about specifying the full parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
inGamma={castable}
specifies the initial values for the shape and rate parameters of the normal-gamma distribution.
For more information about specifying the inGamma parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
inMean={castable}
specifies the initial values of means by using a column-oriented table.
For more information about specifying the inMean parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
mean=double
specifies the default initial value of all means.
| Default | 0 |
|---|
outVerification={checkInitialDistStmt}
outputs the initial distribution to tables for validation.
| Alias | outVerify |
|---|
The checkInitialDistStmt value can be one or more of the following:
covariance={casouttable}
specifies the output table for the covariances of the initial distribution.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
gamma={casouttable}
specifies the output table for the shape and rate parameters of the initial distribution.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
mean={casouttable}
specifies the output table for the means of the initial distribution.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
rate={rateStmt}
specifies the starting setting for the initial distribution's rate parameter.
The rateStmt value can be one or more of the following:
defaultValue=double
specifies the default rate parameter value for all variables.
| Alias | def |
|---|---|
| Default | 5 |
| Minimum value (exclusive) | 0 |
positions={integer-1 <, integer-2, ...>}
specifies the positions of variables for assigning rate parameter values different from the default.
| Alias | pos |
|---|
values={double-1 <, double-2, ...>}
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
shape={shapeStmt}
specifies the starting setting for the initial distribution's shape parameter.
The shapeStmt value can be one or more of the following:
defaultValue=double
specifies the default shape parameter value for all variables.
| Alias | def |
|---|---|
| Default | 10 |
| Minimum value (exclusive) | 1 |
positions={integer-1 <, integer-2, ...>}
specifies the positions of variables for assigning shape parameter values different from the default.
| Alias | pos |
|---|
values={double-1 <, double-2, ...>}
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
model={dlmModelStmt}
specifies the options that are related to the model specification.
* variables={"variable-name-1" <, "variable-name-2", ...>}
specifies the variables to use for the model.
| Alias | vars |
|---|
nSimulations=integer
specifies the number of simulations in filtering and forecasting.
| Alias | nSim |
|---|---|
| Default | 10000 |
| Minimum value | 1 |
parentalSet={dlmParSetStmt}
specifies the model's parental sets.
The dlmParSetStmt value can be one or more of the following:
inAdjacency={castable}
specifies the adjacency matrix of the parental sets.
For more information about specifying the inAdjacency parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | inAdj |
|---|
outVerification={casouttable}
specifies the output table to use for verifying the input adjacency matrices.
For more information about specifying the outVerification parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outVerify |
|---|
seed=64-bit-integer
specifies a nonnegative integer to use as the seed for generating random number sequences.
| Default | 0 |
|---|---|
| Minimum value | 0 |
dynamicLinear Action
Learns and infers dynamic linear models.
Summary: Input and Output Tables
If a row includes a subparameter, you can specify the name, caslib, and so on in the subparameter. Otherwise, you can specify the name, caslib, and so on in the parameter.
|
Parameter |
Subparameter |
Description |
|---|---|---|
|
required parameterdata |
— |
specifies the input data table. |
|
inDiscountFactor |
specifies the model's discount factors. | |
|
inMean, inCovariance (and nested parameter diagonal), (and nested parameter full), inGamma |
specifies the parameters for the initial distribution. | |
|
inAdjacency |
specifies the model's parental sets. |
|
Parameter |
Subparameter |
Description |
|---|---|---|
|
outVerification |
specifies the model's discount factors. | |
|
out, outSim, outCovariance, outFilterInfo (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma), (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma), (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma), (and nested parameter klDivergence), (and nested parameter theta), (and nested parameter nu), (and nested parameter lambda), (and nested parameter theta), (and nested parameter lambda) |
specifies the options that are related to the filtering problem, which aims to find the posterior distribution of the current state, based on observations that precede and include the current observation. | |
|
out, outSim, outCovariance |
specifies the options that are related to the forecasting problem, which aims to determine the probability distribution of future states. | |
|
outVerification (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma) |
specifies the parameters for the initial distribution. | |
|
outVerification |
specifies the model's parental sets. |
Parameter Descriptions
* data={castable}
specifies the input data table.
For more information about specifying the data parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | table |
|---|
discountFactor={dlmDfFactorStmt}
specifies the model's discount factors.
The dlmDfFactorStmt value can be one or more of the following:
"beta":{betaStmt}
specifies the model's stochastic volatility discount factors.
The betaStmt value can be one or more of the following:
"defaultValue":double
specifies the default beta parameter value for all variables.
| Alias | def_ |
|---|---|
| Default | 0.95 |
| Range | (0, 1) |
"positions":[integer-1 <, integer-2, ...>]
specifies the positions of variables for assigning beta parameter values different from the default.
| Alias | pos |
|---|
"values":[double-1 <, double-2, ...>]
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
"deltaGamma":{deltaGammaStmt}
specifies the model's discount factors for regressors in the parental sets.
The deltaGammaStmt value can be one or more of the following:
"defaultValue":double
specifies the default deltaGamma parameter value for variables in the parental sets.
| Alias | def_ |
|---|---|
| Default | 0.995 |
| Range | (0, 1) |
"positions":[integer-1 <, integer-2, ...>]
specifies the positions of variables for assigning deltaGamma parameter values different from the default.
| Alias | pos |
|---|
"values":[double-1 <, double-2, ...>]
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
"deltaPhi":{deltaPhiStmt}
specifies the model's discount factors for regressors not in the parental sets.
The deltaPhiStmt value can be one or more of the following:
"defaultValue":double
specifies the default deltaPhi parameter value for regressors not in the parental sets.
| Alias | def_ |
|---|---|
| Default | 0.995 |
| Range | (0, 1) |
"positions":[integer-1 <, integer-2, ...>]
specifies the positions of variables for assigning deltaPhi parameter values different from the default.
| Alias | pos |
|---|
"values":[double-1 <, double-2, ...>]
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
"inDiscountFactor":{castable}
specifies the initial values for all discount factors.
For more information about specifying the inDiscountFactor parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | inDF |
|---|
"outVerification":{casouttable}
specifies the table to use for validating the input discount factors.
For more information about specifying the outVerification parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outVerify |
|---|
filter={dlmFilterStmt}
specifies the options that are related to the filtering problem, which aims to find the posterior distribution of the current state, based on observations that precede and include the current observation.
The dlmFilterStmt value can be one or more of the following:
"back":integer
specifies the number of observations before the end of the data at which the forecasts begin.
| Default | 1 |
|---|---|
| Minimum value | 0 |
"lead":integer
specifies the number of the multistep forecast values to compute during filtering.
| Default | 1 |
|---|---|
| Minimum value | 1 |
"out":{casouttable}
writes in-sample forecasting results and statistics to an output table.
For more information about specifying the out parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"outCovariance":{casouttable}
writes the covariances of samples of in-sample forecasting observations to an output table.
For more information about specifying the outCovariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outCov |
|---|
"outFilterInfo":{outFilterInfoStmt}
specifies which filtering information to write to output tables.
The outFilterInfoStmt value can be one or more of the following:
"klDivergence":{casouttable}
specifies the output table for Kullback-Leibler divergence information in the decoupling step of updating the distribution from naive posterior to posterior.
For more information about specifying the klDivergence parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | klDiv |
|---|
"naivePosterior":{filterNaivePostStmt}
specifies which parameters of the naive posterior distribution to output.
| Alias | naivePost |
|---|
The filterNaivePostStmt value can be one or more of the following:
"covariance":{casouttable}
specifies the output table for the covariances of naive posterior normal-gamma distributions.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
"gamma":{casouttable}
specifies the output table for the shape and rate parameters of naive posterior normal-gamma distributions.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"mean":{casouttable}
specifies the output table for the means of naive posterior normal-gamma distributions.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"posterior":{filterPostStmt}
specifies which parameters of the posterior distribution to output.
| Alias | post |
|---|
The filterPostStmt value can be one or more of the following:
"covariance":{casouttable}
specifies the output table for the covariances of posterior normal-gamma distributions.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
"gamma":{casouttable}
specifies the output table for the shape and rate parameters of posterior normal-gamma distributions.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"mean":{casouttable}
specifies the output table for the means of posterior normal-gamma distributions.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"predictionSample":{filterPredRawSmplStmt}
specifies which raw random samples to output during forecasting.
| Alias | predSmpl |
|---|
The filterPredRawSmplStmt value can be one or more of the following:
"lambda_":{casouttable}
specifies the output table for raw random samples of precisions during forecasting.
For more information about specifying the lambda parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"nu":{casouttable}
specifies the output table for raw random samples of observation errors during forecasting.
For more information about specifying the nu parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"theta":{casouttable}
specifies the output table for raw random samples of state vectors during forecasting.
For more information about specifying the theta parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"prior":{filterPriorStmt}
specifies which parameters of the prior distribution to output.
The filterPriorStmt value can be one or more of the following:
"covariance":{casouttable}
specifies the output table for the covariances of prior normal-gamma distributions.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
"gamma":{casouttable}
specifies the output table for the shape and rate parameters of prior normal-gamma distributions.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"mean":{casouttable}
specifies the output table for the means of prior normal-gamma distributions.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"recouplingSample":{filterImpSmplRawSmplStmt}
specifies which raw random samples to output during the recoupling of importance sampling.
| Alias | recSmpl |
|---|
The filterImpSmplRawSmplStmt value can be one or more of the following:
"lambda_":{casouttable}
specifies the output table for raw random samples of precisions during the recoupling of importance sampling.
For more information about specifying the lambda parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"theta":{casouttable}
specifies the output table for raw random samples of state vectors during the recoupling of importance sampling.
For more information about specifying the theta parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"outSim":{casouttable}
writes samples of in-sample forecasting observations to an output table.
For more information about specifying the outSim parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"percent":[double-1 <, double-2, ...>]
specifies percentile points for the summary of filtering results.
| Alias | pct |
|---|
forecast={dlmForecastStmt}
specifies the options that are related to the forecasting problem, which aims to determine the probability distribution of future states.
The dlmForecastStmt value can be one or more of the following:
"lead":integer
specifies the number of the multistep forecast values to compute during forecasting.
| Default | 1 |
|---|---|
| Minimum value | 1 |
"out":{casouttable}
writes out-of-sample forecasting results and statistics to an output table.
For more information about specifying the out parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"outCovariance":{casouttable}
writes the covariances of samples of out-of-sample forecasting observations to an output table.
For more information about specifying the outCovariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outCov |
|---|
"outSim":{casouttable}
writes samples of out-of-sample forecasting observations to an output table.
For more information about specifying the outSim parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"percent":[double-1 <, double-2, ...>]
specifies percentile points for the summary of forecasting results.
| Alias | pct |
|---|
* id="variable-name"
identifies observations in the input data table by specifying a variable for the time series data.
initialDistribution={dlmInitDistStmt}
specifies the parameters for the initial distribution.
The dlmInitDistStmt value can be one or more of the following:
"inCovariance":{covStmt}
specifies the starting covariance settings for the initial distribution.
| Alias | inCov |
|---|
The covStmt value can be one or more of the following:
"diagonal":{castable}
specifies the initial variance values (diagonal) for the initial distribution's covariances.
For more information about specifying the diagonal parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | diag |
|---|
"full":{castable}
specifies the initial covariance values by using a row-oriented table.
For more information about specifying the full parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
"inGamma":{castable}
specifies the initial values for the shape and rate parameters of the normal-gamma distribution.
For more information about specifying the inGamma parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
"inMean":{castable}
specifies the initial values of means by using a column-oriented table.
For more information about specifying the inMean parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
"mean":double
specifies the default initial value of all means.
| Default | 0 |
|---|
"outVerification":{checkInitialDistStmt}
outputs the initial distribution to tables for validation.
| Alias | outVerify |
|---|
The checkInitialDistStmt value can be one or more of the following:
"covariance":{casouttable}
specifies the output table for the covariances of the initial distribution.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
"gamma":{casouttable}
specifies the output table for the shape and rate parameters of the initial distribution.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"mean":{casouttable}
specifies the output table for the means of the initial distribution.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
"rate":{rateStmt}
specifies the starting setting for the initial distribution's rate parameter.
The rateStmt value can be one or more of the following:
"defaultValue":double
specifies the default rate parameter value for all variables.
| Alias | def_ |
|---|---|
| Default | 5 |
| Minimum value (exclusive) | 0 |
"positions":[integer-1 <, integer-2, ...>]
specifies the positions of variables for assigning rate parameter values different from the default.
| Alias | pos |
|---|
"values":[double-1 <, double-2, ...>]
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
"shape":{shapeStmt}
specifies the starting setting for the initial distribution's shape parameter.
The shapeStmt value can be one or more of the following:
"defaultValue":double
specifies the default shape parameter value for all variables.
| Alias | def_ |
|---|---|
| Default | 10 |
| Minimum value (exclusive) | 1 |
"positions":[integer-1 <, integer-2, ...>]
specifies the positions of variables for assigning shape parameter values different from the default.
| Alias | pos |
|---|
"values":[double-1 <, double-2, ...>]
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
model={dlmModelStmt}
specifies the options that are related to the model specification.
* "variables":["variable-name-1" <, "variable-name-2", ...>]
specifies the variables to use for the model.
| Alias | vars |
|---|
nSimulations=integer
specifies the number of simulations in filtering and forecasting.
| Alias | nSim |
|---|---|
| Default | 10000 |
| Minimum value | 1 |
parentalSet={dlmParSetStmt}
specifies the model's parental sets.
The dlmParSetStmt value can be one or more of the following:
"inAdjacency":{castable}
specifies the adjacency matrix of the parental sets.
For more information about specifying the inAdjacency parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | inAdj |
|---|
"outVerification":{casouttable}
specifies the output table to use for verifying the input adjacency matrices.
For more information about specifying the outVerification parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outVerify |
|---|
seed=64-bit-integer
specifies a nonnegative integer to use as the seed for generating random number sequences.
| Default | 0 |
|---|---|
| Minimum value | 0 |
dynamicLinear Action
Learns and infers dynamic linear models.
Summary: Input and Output Tables
If a row includes a subparameter, you can specify the name, caslib, and so on in the subparameter. Otherwise, you can specify the name, caslib, and so on in the parameter.
|
Parameter |
Subparameter |
Description |
|---|---|---|
|
required parameterdata |
— |
specifies the input data table. |
|
inDiscountFactor |
specifies the model's discount factors. | |
|
inMean, inCovariance (and nested parameter diagonal), (and nested parameter full), inGamma |
specifies the parameters for the initial distribution. | |
|
inAdjacency |
specifies the model's parental sets. |
|
Parameter |
Subparameter |
Description |
|---|---|---|
|
outVerification |
specifies the model's discount factors. | |
|
out, outSim, outCovariance, outFilterInfo (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma), (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma), (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma), (and nested parameter klDivergence), (and nested parameter theta), (and nested parameter nu), (and nested parameter lambda), (and nested parameter theta), (and nested parameter lambda) |
specifies the options that are related to the filtering problem, which aims to find the posterior distribution of the current state, based on observations that precede and include the current observation. | |
|
out, outSim, outCovariance |
specifies the options that are related to the forecasting problem, which aims to determine the probability distribution of future states. | |
|
outVerification (and nested parameter mean), (and nested parameter covariance), (and nested parameter gamma) |
specifies the parameters for the initial distribution. | |
|
outVerification |
specifies the model's parental sets. |
Parameter Descriptions
* data=list(castable)
specifies the input data table.
For more information about specifying the data parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | table |
|---|
discountFactor=list(dlmDfFactorStmt)
specifies the model's discount factors.
The dlmDfFactorStmt value can be one or more of the following:
beta=list(betaStmt)
specifies the model's stochastic volatility discount factors.
The betaStmt value can be one or more of the following:
defaultValue=double
specifies the default beta parameter value for all variables.
| Alias | def |
|---|---|
| Default | 0.95 |
| Range | (0, 1) |
positions=list(integer-1 <, integer-2, ...>)
specifies the positions of variables for assigning beta parameter values different from the default.
| Alias | pos |
|---|
values=list(double-1 <, double-2, ...>)
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
deltaGamma=list(deltaGammaStmt)
specifies the model's discount factors for regressors in the parental sets.
The deltaGammaStmt value can be one or more of the following:
defaultValue=double
specifies the default deltaGamma parameter value for variables in the parental sets.
| Alias | def |
|---|---|
| Default | 0.995 |
| Range | (0, 1) |
positions=list(integer-1 <, integer-2, ...>)
specifies the positions of variables for assigning deltaGamma parameter values different from the default.
| Alias | pos |
|---|
values=list(double-1 <, double-2, ...>)
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
deltaPhi=list(deltaPhiStmt)
specifies the model's discount factors for regressors not in the parental sets.
The deltaPhiStmt value can be one or more of the following:
defaultValue=double
specifies the default deltaPhi parameter value for regressors not in the parental sets.
| Alias | def |
|---|---|
| Default | 0.995 |
| Range | (0, 1) |
positions=list(integer-1 <, integer-2, ...>)
specifies the positions of variables for assigning deltaPhi parameter values different from the default.
| Alias | pos |
|---|
values=list(double-1 <, double-2, ...>)
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
inDiscountFactor=list(castable)
specifies the initial values for all discount factors.
For more information about specifying the inDiscountFactor parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | inDF |
|---|
outVerification=list(casouttable)
specifies the table to use for validating the input discount factors.
For more information about specifying the outVerification parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outVerify |
|---|
filter=list(dlmFilterStmt)
specifies the options that are related to the filtering problem, which aims to find the posterior distribution of the current state, based on observations that precede and include the current observation.
The dlmFilterStmt value can be one or more of the following:
back=integer
specifies the number of observations before the end of the data at which the forecasts begin.
| Default | 1 |
|---|---|
| Minimum value | 0 |
lead=integer
specifies the number of the multistep forecast values to compute during filtering.
| Default | 1 |
|---|---|
| Minimum value | 1 |
out=list(casouttable)
writes in-sample forecasting results and statistics to an output table.
For more information about specifying the out parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
outCovariance=list(casouttable)
writes the covariances of samples of in-sample forecasting observations to an output table.
For more information about specifying the outCovariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outCov |
|---|
outFilterInfo=list(outFilterInfoStmt)
specifies which filtering information to write to output tables.
The outFilterInfoStmt value can be one or more of the following:
klDivergence=list(casouttable)
specifies the output table for Kullback-Leibler divergence information in the decoupling step of updating the distribution from naive posterior to posterior.
For more information about specifying the klDivergence parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | klDiv |
|---|
naivePosterior=list(filterNaivePostStmt)
specifies which parameters of the naive posterior distribution to output.
| Alias | naivePost |
|---|
The filterNaivePostStmt value can be one or more of the following:
covariance=list(casouttable)
specifies the output table for the covariances of naive posterior normal-gamma distributions.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
gamma=list(casouttable)
specifies the output table for the shape and rate parameters of naive posterior normal-gamma distributions.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
mean=list(casouttable)
specifies the output table for the means of naive posterior normal-gamma distributions.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
posterior=list(filterPostStmt)
specifies which parameters of the posterior distribution to output.
| Alias | post |
|---|
The filterPostStmt value can be one or more of the following:
covariance=list(casouttable)
specifies the output table for the covariances of posterior normal-gamma distributions.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
gamma=list(casouttable)
specifies the output table for the shape and rate parameters of posterior normal-gamma distributions.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
mean=list(casouttable)
specifies the output table for the means of posterior normal-gamma distributions.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
predictionSample=list(filterPredRawSmplStmt)
specifies which raw random samples to output during forecasting.
| Alias | predSmpl |
|---|
The filterPredRawSmplStmt value can be one or more of the following:
lambda=list(casouttable)
specifies the output table for raw random samples of precisions during forecasting.
For more information about specifying the lambda parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
nu=list(casouttable)
specifies the output table for raw random samples of observation errors during forecasting.
For more information about specifying the nu parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
theta=list(casouttable)
specifies the output table for raw random samples of state vectors during forecasting.
For more information about specifying the theta parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
prior=list(filterPriorStmt)
specifies which parameters of the prior distribution to output.
The filterPriorStmt value can be one or more of the following:
covariance=list(casouttable)
specifies the output table for the covariances of prior normal-gamma distributions.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
gamma=list(casouttable)
specifies the output table for the shape and rate parameters of prior normal-gamma distributions.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
mean=list(casouttable)
specifies the output table for the means of prior normal-gamma distributions.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
recouplingSample=list(filterImpSmplRawSmplStmt)
specifies which raw random samples to output during the recoupling of importance sampling.
| Alias | recSmpl |
|---|
The filterImpSmplRawSmplStmt value can be one or more of the following:
lambda=list(casouttable)
specifies the output table for raw random samples of precisions during the recoupling of importance sampling.
For more information about specifying the lambda parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
theta=list(casouttable)
specifies the output table for raw random samples of state vectors during the recoupling of importance sampling.
For more information about specifying the theta parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
outSim=list(casouttable)
writes samples of in-sample forecasting observations to an output table.
For more information about specifying the outSim parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
percent=list(double-1 <, double-2, ...>)
specifies percentile points for the summary of filtering results.
| Alias | pct |
|---|
forecast=list(dlmForecastStmt)
specifies the options that are related to the forecasting problem, which aims to determine the probability distribution of future states.
The dlmForecastStmt value can be one or more of the following:
lead=integer
specifies the number of the multistep forecast values to compute during forecasting.
| Default | 1 |
|---|---|
| Minimum value | 1 |
out=list(casouttable)
writes out-of-sample forecasting results and statistics to an output table.
For more information about specifying the out parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
outCovariance=list(casouttable)
writes the covariances of samples of out-of-sample forecasting observations to an output table.
For more information about specifying the outCovariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outCov |
|---|
outSim=list(casouttable)
writes samples of out-of-sample forecasting observations to an output table.
For more information about specifying the outSim parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
percent=list(double-1 <, double-2, ...>)
specifies percentile points for the summary of forecasting results.
| Alias | pct |
|---|
* id="variable-name"
identifies observations in the input data table by specifying a variable for the time series data.
initialDistribution=list(dlmInitDistStmt)
specifies the parameters for the initial distribution.
The dlmInitDistStmt value can be one or more of the following:
inCovariance=list(covStmt)
specifies the starting covariance settings for the initial distribution.
| Alias | inCov |
|---|
The covStmt value can be one or more of the following:
diagonal=list(castable)
specifies the initial variance values (diagonal) for the initial distribution's covariances.
For more information about specifying the diagonal parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | diag |
|---|
full=list(castable)
specifies the initial covariance values by using a row-oriented table.
For more information about specifying the full parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
inGamma=list(castable)
specifies the initial values for the shape and rate parameters of the normal-gamma distribution.
For more information about specifying the inGamma parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
inMean=list(castable)
specifies the initial values of means by using a column-oriented table.
For more information about specifying the inMean parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
mean=double
specifies the default initial value of all means.
| Default | 0 |
|---|
outVerification=list(checkInitialDistStmt)
outputs the initial distribution to tables for validation.
| Alias | outVerify |
|---|
The checkInitialDistStmt value can be one or more of the following:
covariance=list(casouttable)
specifies the output table for the covariances of the initial distribution.
For more information about specifying the covariance parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | cov |
|---|
gamma=list(casouttable)
specifies the output table for the shape and rate parameters of the initial distribution.
For more information about specifying the gamma parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
mean=list(casouttable)
specifies the output table for the means of the initial distribution.
For more information about specifying the mean parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
rate=list(rateStmt)
specifies the starting setting for the initial distribution's rate parameter.
The rateStmt value can be one or more of the following:
defaultValue=double
specifies the default rate parameter value for all variables.
| Alias | def |
|---|---|
| Default | 5 |
| Minimum value (exclusive) | 0 |
positions=list(integer-1 <, integer-2, ...>)
specifies the positions of variables for assigning rate parameter values different from the default.
| Alias | pos |
|---|
values=list(double-1 <, double-2, ...>)
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
shape=list(shapeStmt)
specifies the starting setting for the initial distribution's shape parameter.
The shapeStmt value can be one or more of the following:
defaultValue=double
specifies the default shape parameter value for all variables.
| Alias | def |
|---|---|
| Default | 10 |
| Minimum value (exclusive) | 1 |
positions=list(integer-1 <, integer-2, ...>)
specifies the positions of variables for assigning shape parameter values different from the default.
| Alias | pos |
|---|
values=list(double-1 <, double-2, ...>)
specifies the values that correspond to the positions parameter list.
| Alias | val |
|---|
model=list(dlmModelStmt)
specifies the options that are related to the model specification.
* variables=list("variable-name-1" <, "variable-name-2", ...>)
specifies the variables to use for the model.
| Alias | vars |
|---|
nSimulations=integer
specifies the number of simulations in filtering and forecasting.
| Alias | nSim |
|---|---|
| Default | 10000 |
| Minimum value | 1 |
parentalSet=list(dlmParSetStmt)
specifies the model's parental sets.
The dlmParSetStmt value can be one or more of the following:
inAdjacency=list(castable)
specifies the adjacency matrix of the parental sets.
For more information about specifying the inAdjacency parameter, see the common castable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | inAdj |
|---|
outVerification=list(casouttable)
specifies the output table to use for verifying the input adjacency matrices.
For more information about specifying the outVerification parameter, see the common casouttable (Form 2) parameter (Appendix A: Common Parameters).
| Alias | outVerify |
|---|
seed=64-bit-integer
specifies a nonnegative integer to use as the seed for generating random number sequences.
| Default | 0 |
|---|---|
| Minimum value | 0 |