Overview (SIMPLAN command)

The SIMPLAN command creates a simulation plan that contains all of the specifications required to run a simulation. The simulation plan is used as input to the SIMRUN command, which runs the simulation.

Simulation in IBM® SPSS® Statistics refers to simulating input data to predictive models, evaluating the model based on the simulated data and using the distribution of predicted target values to evaluate the likelihood of various outcomes. Inputs to the model are specified as fixed or simulated. Values for simulated inputs are drawn from a specified probability distribution. Fixed inputs are those whose values are known and remain constant when evaluating the model. The simulation plan specifies the model to be simulated, probability distributions for simulated inputs and values for fixed inputs, as well as various other settings.

You can also simulate data in the absence of a predictive model by specifying probability distributions for variables that are to be simulated. The simulation plan then specifies those probability distributions, as well as various other settings.

Options

Probability distributions for simulated inputs. You can automatically determine the probability distribution that most closely fits the data, in the active dataset, associated with each simulated input. You can also manually specify distributions if historical data are not available or you require specific distributions.

Correlations. You can automatically determine correlations between simulated inputs from historical data for each input in the active dataset. You can also manually specify correlations between simulated inputs if historical data are not available or you require specific correlations.

Associations for categorical inputs. You can automatically compute a multiway contingency table, from the historical data, that describes the associations between inputs with a categorical distribution. The contingency table is then used when generating data for those inputs.

Input and target filtering. You can specify the valid range of values for both simulated inputs and targets. For simulated inputs, only values in the specified range will be used in simulated data. For targets, values outside of the specified range are excluded from all output and analyses.

Specify sensitivity analysis. Sensitivity analysis allows you to investigate the effect of systematic changes in a fixed input or in a distribution parameter for a simulated input by generating an independent set of simulated data—effectively, a separate simulation—for each specified value.

Stopping criteria. You can specify criteria for stopping the simulation, potentially before the maximum number of allowable cases of simulated data have been generated. In particular, you can specify to continue generating cases until the confidence interval of the mean of a target is within a specified threshold, or you can specify to continue generating cases until one of the tails of a target has been sampled to a specified degree.

Creating a simulation plan that includes a predictive model

The basic specification is the SIMPLAN command with the CREATE keyword, followed by the MODEL subcommand, one or more SIMINPUT subcommands for simulated inputs and the PLAN subcommand specifying where the simulation plan should be saved.

Creating a simulation plan without a predictive model

The basic specification is the SIMPLAN command with the CREATE keyword, followed by one or more SIMINPUT subcommands for simulated inputs and the PLAN subcommand specifying where the simulation plan should be saved.

Displaying an existing plan

The basic specification is the SIMPLAN command with the VIEW keyword, followed by the PLAN subcommand which specifies the simulation plan to be displayed.

Syntax rules

  • The PLAN subcommand, and at least one SIMINPUT subcommand are required. All other subcommands are optional.
  • Subcommands can be specified in any order.
  • There can be multiple instances of SIMINPUT, FIXEDINPUT, TARGETOPTS, and VALUELABELS subcommands. Only a single instance of any other subcommand is allowed.
  • A keyword cannot be specified more than once within a subcommand.
  • Only one SIMINPUT subcommand can specify iterations of distribution parameters.
  • If multiple input fields are specified for a SIMINPUT subcommand, then you cannot specify iterations of distribution parameters for that subcommand.
  • Only one FIXEDINPUT subcommand can contain the ITERATIONS keyword.
  • If multiple input fields are specified for a FIXEDINPUT subcommand, then the subcommand cannot contain the ITERATIONS keyword.
  • If a FIXEDINPUT subcommand contains the ITERATIONS keyword, then iterations of distribution parameters cannot be specified for a SIMINPUT subcommand.
  • Parentheses, equals signs, and slashes shown in the syntax chart are required.
  • The command name, subcommand names, and keywords must be spelled in full.
  • Empty subcommands are not allowed.

Limitations

  • SPLIT FILE is ignored by this command.