The effort documented in this report includes an analysis of historical data related to government R&D program cost overruns and a preliminary assessment of the implications of such analysis for management reserves for related high-risk DARPA programs. This assessment will ultimately lead to the specification of a process to guide DARPA management in the establishment of budgetary reserves to account for technological risk and to periodically update projections of selected program cost growth based on experience. The key to the work performed so far under this contract is the establishment of a data base containing information on historical cost growth of major government R&D and production programs during the period FY 1977 through FY 1980. Based on statistical analysis of these data subsets, trends relating probabilistic measures of cost growth and uncertainty were developed. The analysis shows that based on historical evidence, cost overruns of significant size (from 50 to 100 percent) are not uncommon. In addition the analysis shows that there is a predictable relationship between the size of an original estimate of program overrun (i.e., management reserve) and the confidence level that eventual cost growth will remain within that limit. For the initial set of DARPA programs studied, the analysis indicates that, at the 95 percent confidence limit, DARPA cost growth factors (i.e., the ratio of actual programs costs to original estimates) lie in the range of approximately 40 to 110 percent.