*1959*

**RAND Corporation**

“Results of some recent research into the extent and nature of the uncertainty in new developments, with emphasis on problems of development in the Air Force. “Early” estimates of important parameters are usually quite inaccurate because they are “biased” toward overoptimism and because the errors in estimates evidence a substantial variation. The accuracy of estimates is found to be a function of the stage of development, i.e., estimates improve as development of the item progresses.”

*1958*

**RAND Corporation**

“A discussion of simulation and Monte Carlo as modes of analysis of particular interest in problems of operations analysis involving many variables. After defining these terms, the author considers (1) Monte Carlo design when simulation aspects are not emphasized, and (2) simulation design with Monte Carlo aspects.”

*1957*

**RAND Corporation**

“A discussion of asymptotic distribution. (Published in The Annals of Mathematical Statistics, Mar. 1958.)”

*1955*

**RAND Corporation**

“A discussion of (1) the progress of Monte Carlo techniques from 1949 to the present and (2) the papers presented before the Symposium on Monte Carlo Methods at Gainesvill, Florida, March 16-17, 1954. Monte Carlo is defined as a random sampling procedure for treating mathematical problems of either deterministic or probabilistic type.”

*1955*

**RAND Corporation**

“A discussion of the operation of randomized branch sampling, the relation between this sampling plan and importance sampling developed in Monte Carlo calculations, and some improvements to Jessen’s theory based on the identity of these methods.”

*1954*

**RAND Corporation**

“A presentation of several methods (developed in studies of mental disease) for determining certain epidemiological parameters that are not directly observable or that can be secured only by expensive and time-consuming field surveys. The simple models of the process involved in the passage from sanity to insanity, hospitalization and death provide some picture of the underlying process that generates a given incidence rate.”

*1954*

**RAND Corporation**

“A review of importance sampling, a technique whereby Monte Carlo computations are made more efficient — provided the probability distribution, from which the sample observations are drawn, is judiciously chosen. (Published in H. A. Meyer (ed.), Symposium on Monte Carlo Methods, John Wiley & Sons, Inc., New York, 1956.)”

*1953*

**RAND Corporation**

“This paper formulates the general problem of Monte Carlo computations within the framework of mathematical statistics and indicates briefly its relation to the theory of estimation and the design of experiments. (Published in the Journal of the Operations Research Society of America, Nov. 1953.)”

*1951*

**RAND Corporation**

“A discussion of the appropriateness of comparisons presented in P-208, “Specific Industry Output Projections”, for decisions as to the acceptance or rejection of proposed forecasting methods. An alternative method of measuring the forecasting errors is suggested.”