Abstract :
Summary form only given. A new generation of software performs Monte Carlo simulation nearly instantaneously on arbitrary spreadsheet models. In effect, it does for probability distributions what the spreadsheet did for numbers. In addition, the concept of a stochastic information system is introduced, in which distributions are stored as realizations of a stochastic process. A single variable is stored as a stochastic information packet (SIP), a multivariate distribution as a stochastic library unit, relationships preserved (SLURP). This leads to the fundamental equality of SLURP algebra: P(F(x,y)) = F(P(x,y)) where x and y are random variables, p(x,y) is the SLURP representing their joint distribution, and F is a function of x and y. Live computer demonstrations include distribution arithmetic, the flaw of averages, a retirement simulation developed for a New York trust company, and interactive portfolio model used at a major petroleum firm
Keywords :
Monte Carlo methods; digital simulation; spreadsheet programs; Monte Carlo simulation; arbitrary spreadsheet models; distribution arithmetic; fundamental equality; interactive portfolio model; interactive simulation; multivariate distribution; probability distributions; retirement simulation; stochastic information system; stochastic library unit; Algebra; Digital arithmetic; Distributed computing; Information systems; Libraries; Probability distribution; Random variables; Software performance; Stochastic processes; Stochastic systems;
