PENSIM input parameters

PENSIM input table RUN

Contains highest level description of model runs, consisting mostly of foreign keys pointing to rows in child tables. Each row specifies a different run.

Parent table is: QUEUE

Child table is: OUTPUT

Child table is: INDBEHV

Child table is: INDBEHX

Child table is: PENCHAR

Child table is: ASSETS

Child table is: ACCTFEE

Child table is: PENBEHV

Child table is: PENLAW

Child table is: ECI

Child table is: INCTAX

Go to PENSIM input parameters overview

PENSIM: RUN . id
Identifies run specified on this row of table, so id value should be unique.

Valid values: 1 to 99999, Integer

PENSIM: RUN . notes
Describes specification of table row in free-form note.

Valid values: any ASCII text, Memo (Note: do not copy and paste from a word processor because of the possibility of embedded non-ASCII characters.)

PENSIM: RUN . ssasim_dir
Specifies database subdirectory/subfolder in which the SSASIM run used by this PENSIM run is located. This SSASIM run specifies the macroeconomic and macrodemographic environment within which the cohort sample lives.

Valid values: enter any valid subdirectory/subfolder of the SSASIM executables directory/folder (specified in the %ssasim% environment variable) without the trailing backslash; embedded spaces in subdirectory/subfolder name not allowed.

PENSIM: RUN . ssasim_rid
Specifies RUN.id value of the SSASIM run used by this PENSIM run. This SSASIM run specifies the macroeconomic and macrodemographic environment within which the cohort sample lives.

Valid values: 1 to 99999, Integer

PENSIM: RUN . birth_year
Specifies calendar year in which cohort sample individuals are born.

Valid values: 1935 to 2934, Integer

PENSIM: RUN . scenarios
Specifies total number of stochastic scenarios used in this run. Typical values range from 1 (for a run that assumes a certain or deterministic environment for the cohort individuals) to 1000 (for a run that assumes an uncertain or stochastic environment for the cohort individuals). Typically, PENSIM:RUN.scenarios will equal the value of SSASIM:RUN.scenarios for the run specified PENSIM:RUN.ssasim_dir/rid. In this typical case, the number of scenarios in the cohort environment (env) file produced by SSASIM for PENSIM will be equal to the number of PENSIM run scenarios. However, in one special case, the value of PENSIM:RUN.scenarios may be set to one (1) while the value of SSASIM:RUN.scenarios is set to be greater than one (1). This special case is one in which no demographic or economic variables (other than asset returns and the interest rate) vary across SSASIM scenarios. In this case the PENSIM no-pension life histories generated for GEMINI input will be the same across scenarios, so there is no need to generate a scenario sample more than once. Using the capability provided in this special case will reduce GEMINI and SSASIM OLC mode execution times for runs in which one asset returns are varying across scenarios (as may be the case in stochastic policy analysis of individual account reforms).

Valid values: 1 to 99999, Integer [typical range is 1 to 1000]

PENSIM: RUN . sample_pct
Specifies the cohort scenario sampling percent. Note that this is a per-scenario sampling percent. The cohort sample size for each scenario is calculated by multiplying this specified percent by the number of native births and immigrants specified in the cohort environment (envNNNNN) file produced by the SSASIM run specified in the RUN.ssasim_dir and RUN.ssasim_rid fields. Note that the sum of native births and immigrants for a birth cohort may vary across scenarios if the SSASIM run specifies stochastic fertility rates or immigration flows. The use of a sampling percent (rather than decimal fraction) means that sample_pct=0.1 produces a one-in-a-thousand sample of the real-world cohort.

Valid values: 0.0001 to 10.0, Real

PENSIM: RUN . noweddings
Specifies whether or not cohort sample individuals get married. If this parameter is false, it has no effect on the run with marriages being simulated as specified in the WEDDING and child tables. If this parameter is true, then no weddings are simulated in the run because the values of WEDDING.last_age_f and WEDDING.last_age_m are both set to zero, overriding the values of those two parameters specified in the WEDDING table.

Valid values: T (=1) or F (=0) (for true or false), Logical

PENSIM: RUN . max_s_list
Specifies the maximum length of each potential spouse list, of which there are many. This is a technical parameter that controls how many spouse life histories are saved for future weddings, in cases where the simulated spouse is not a suitable marriage partner for the cohort individual currently being married (because of inappropriate schooling or martial status), but may be suitable for a different cohort individual. These spouse lists are purged a the end of each scenario, so that each cohort sample individual and spouse(s) always live in the same environment. Changing the value of this parameter will cause a change in spouses (because the spouse matching process will be different), so never compare runs specify different values for this field. A value of 40 seems to maximize run execution speed.

Valid values: 0 to 99, Integer

PENSIM: RUN . rng_offset
Specifies the number of random numbers drawn when each random number generator is initialized. If changing the value of this parameter produces substantial differences in the summary statistics for the sample generated by this run, then the sample_pct needs to be increased to reduce the Monte Carlo sampling variance. This parameter is not read when PENSIM is called by SSASIM as an engine in support of an EPS mode, ESP+OLC mode, RCS mode, RCS+OLC mode, or OLC mode run. In these cases, the value of SSASIM:RUN.rng_offset is used in place of the value specified in this field.

Valid values: 0 to 999, Integer

PENSIM: RUN . rng_reset
Specifies whether or not the random number generators, which determine individual life histories along with the macro environment, are reset at the start of each scenario. If false, each scenario sample is generated using different random numbers as well as possibly a different macro environment. If true, each scenario sample is generated using the same random numbers, which means the scenario samples will contain the same group of individuals only if there are no differences in the macro environment across scenarios.

Valid values: T (=1) or F (=0) (for true or false), Logical

PENSIM: RUN . output_id
Points to a row in the child OUTPUT table, which specifies the particular output files that will be written by this run.