Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
FUTURE COMPUTING ENVIRONMENTS FOR MICROSIMULATION MODELING 159 Output Facilities As mentioned above, SPSD/M supports a rich set of output facilities. This section provides a consolidated list of the types of output. Built-in tables can be produced optionally during any SPSD/M run. There are five sets of pairs of tables. The first set of each pair contains values in millions of dollars for over 50 selected variables. The second table in each pair contains the number of persons, families, or households that had some nonzero value for the variable reported. The five tables are (1) Canada totals, (2) totals by province, (3) totals by income group, (4) totals by family type, and (5) totals by proportion of specified income threshold. The user has limited control over the format and contents of the built-in tables. The user specifies the unit of analysis that applies to all the tables and can specify the variable as well as the cut-points for the third table. A separate utility program is provided to load the built-in tables into a Lotus 1â2â3 spreadsheet, so that the user can perform subsequent analysis. The X-Tab facility can be used to create custom tables. The X-Tab specification indicates the relevant unit of analysis (e.g., EF indicates economic families), followed by one or more levels separated by asterisks. The tabulation level indicates the items to be tabulated. Any remaining levels are classification levels, which specify the particular pattern of categorization desired in the table. For example, the following X-Tab specification would tabulate total income accruing to economic families, the number of such families, and the average income per family, all classified by tenure category (e.g., rented, owned with mortgage): The distributional analysis facility calculates a set of univariate distributional statistics for a selected variable for a sample of SPSD/M cases. The facility processes a user-specified random sample of up to 10,000 cases and calculates the number of zero and nonzero observations, descriptive statistics such as s um of weights, extreme values, selected quantiles, and a histogram plot. The distributional analysis is output with the built-in and custom tables. The marginal tax rate facility permits SPSD/M to calculate the marginal tax rate (i.e., the proportion of an extra dollar of income that is taxed). The user specifies the income variable to be incremented and the incremental value, and SPSD/M executes the tax/transfer system twice for each householdâonce with the original incomes and once with the incremented incomesâand the resulting change in consumable income is noted in a special variable. The change in individual consumable income can then be aggregated to produce marginal tax rates at different family levels of analysis. Parameter files that document the scenario are output at the end of the model run. These files are created in a standard self-documenting format, with