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 201 that are believable and (2) simulation systems that are versatile and flexible yet relatively inexpensive and easy to use. The rudimentary state of both of these factors is consistent with the paucity of such models and modeling systems today. The production of knowledge in economics and related social science disciplines is a slow and laborious process for which there is no reliable production function. Although specific studies can be performed to increase knowledge at a specific time in a specific area, it is not now practical from either a cost or time standpoint to commission the research required to build a specific microanalytic model with any degree of completeness or closure.55 The substance of models must generally stand on knowledge that is already available. The computer-based component affecting the availability and supply of microanalytic simulation models is more tractable to improvement. Evidence presented in the previous section leads to the strong presumption that there will be substantially more computing power available in the next few years, on the desktop, at current prices. This increase in power can be used in a number of complementary ways, including lowering the costs associated with microsimulation experiments and making the user interface easier to use. The underlying computing industry is experiencing rapid growth, with a high level of technical progress and innovation in hardware and software. We believe that it is possible to achieve major overall productivity and performance advantages by exploiting advances in hardware and software technology and that this should be a major thrust in the development of computer-based systems to support microanalytic simulation in the 1990s. Functions of a System for Microanalytic Simulation Computer programs have certain basic functions they perform in order to accomplish useful work. They accept and check input that defines a particular set of options and input values to be used. They allocate memory âprimary, secondary, or bothâfor data used during the execution of the program. They allocate and use other resources to accomplish the task to be performed. They produce outputs, machine and huma n readable, that provide their reason for existence. A computer program is useful to the extent that it performs these functions in a manner that fits within existing computer system resources, is efficient in 55 An example is provided by the Panel Study of Income Dynamics, administered by the University of Michigan Survey Research Center. This longitudinal survey is providing much useful information regarding the social and economic behavior of individuals and families but at a cost of millions of dollars and many years of data collection and analysis. Results derived from the survey are making a major contribution to microanalytic model construction. In particular, the labor force participation model in DYNASIM2 was estimated from these data.