Pension Projection Models
The PSG models have a number of capabilities that make it easy to specify a model run and easy to analyze output from a run.
With regard to specifying model runs, there is no need to engage in the time-consuming and error-prone process of reprogramming the models because all the assumption and policy parameters used in a model run are stored in a relational database for each model that is read at the start of each run. The PSG models use the popular SQLite relational database.
While a relational database can be manipulated with Structured Query Language, or SQL, the PSG models provide a simple run specification language that describes how a run's assumption and/or policy parameters differ from standard benchmark runs already specified in the database. The specification of each run is stored in a run specification file (rsf) that is a simple ASCII text file.
In order to make run specification files useful, the PSG models also provide an RSF Toolkit, which is an rsf editor that has the ability to automatically write a run specification file given user responses to an interview about the nature of the desired run. The RSF Toolkit has many other features that aid the management and use of a large number of run specification files.
In addition, the PSG models provide a database run compiler that translates a run specification file into all the table rows in the three model databases that are required to fully specify the run. This dbrun-create ability eliminates any need to work directly with the model databases.
With regard to analyzing model output, the PSG models provide many analyzer programs that perform a range of common statistical operations, such as computing the mean and percentile values of an output statistic. These analyzer programs are command-line utilities that can be used in scripts to automate complex output analysis tasks.
Because the output files are tab-delimited text files, they can be easily imported into any statistical package (SAS, SPSS, etc.) for custom analysis. For those who want something faster and cheaper than the statistical packages, the GNU AWK interpreter (gawk.exe) and documentation is included in the PSG models installation. For those more comfortable with spreadsheets, all but the very largest output files can be easily imported into Excel.
For interactive output analysis using a GUI program, the PSG models provide two analysis Toolkits that generate graphic visualizations of output related to social security solvency and to the distribution of retirement income.
Other key features.