Accession Number : ADP002881

Title :   Software for Explicitly Probabilistic Mathematics,

Corporate Author : MITRE CORP MCLEAN VA

Personal Author(s) : Goldstein,S. N.

Report Date : DEC 1983

Pagination or Media Count : 10

Abstract : The conventional way to estimate quantities for decision making purposes is to employ informed judgment about the applicable estimating relationships (equations, algorithms, models) and the values of their constituent parameters. More often than not, one or more key parameters are not known with certainty, and informed judgment takes the form of a subjective point estimate (guesstimate). Many practitioners are content to use the resulting point estimate for the result. The more diligent will re-do the calculation combining estimates of upper and lower bounds of the input parameters to estimates upper and lower limits for the result. Such diligence is facilitated by the increasingly convenient spread sheet computational packages for personal computers. This paper describes software for incorporating uncertainty directly and explicitly into the calculations to yield a probability distribution as the result rather than a set of point estimates. The software permits the user to represent the uncertain parameters by any of about sixteen probability density functions.

Descriptors :   *Computer programs, *Probability distribution functions, *Probability density functions, Decision making, Computations, Command and control systems, Message processing

Distribution Statement : APPROVED FOR PUBLIC RELEASE