Research project cost distributions and budget forecasting

Abstract

The purpose of this paper is to determine the nature of actual research project costs, their probability distributions, and corresponding parameter values so that long-range budget forecasts and variances can be provided. The distribution and parameters of the research project costs for 1963 through 1967 have been developed and are discussed. A description of the present Army budgeting system is presented. The budget forecasting problem is defined and a mathematical model is presented. <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> The essential elements of the model are described in terms of parameters and coefficients, which are obtained from the 1963 through 1967 budget data. Four variations of the basic model are developed and compared. This paper contains a list of computer program statements specifically designed 1) to derive the necessary parameters from historical data, and 2) to forecast these parameters for each of five years into the future. The program, its operation, and features are discussed with sample outputs provided for illustration purposes.