Digital Computer Simulation: Input-Output Analysis

Full Description: FISHMAN, George S. Digital Computer Simulation: Input-Output Analysis. Prepared for: United States Air Force Project Rand. Santa Monica, CA: The Rand Corporation, 1968. Octavo (10 7/8 x 8 5/8 inches; 276 x 213 mm). vii, [1, blank], 25, [3, blank] pp. Publisher's printed tan wrappers. Very minor sunning to back wrapper. About fine. [Together with]: Digital Computer Simulation: Estimating Sample Size. Prepared for: United States Air Force Project Rand.Santa Monica, CA: The Rand Corporation, 1969. Octavo (11 x 8 1/2 inches; 279 x 215 mm). vii, [1, blank], 54, [2, blank] pp. Publisher's printed tan wrappers. Some minor smudging to wrapper. Previous owner's stamp on inside of front wrapper dated 1969. About fine. "[In] 1962, George was offered a full-time position as a research analyst at RAND, and he left Stanford with a master's degree in economics. RAND was an exceptionally stimulating place, and the decision to go there was not hard for George to make. In the early 1960s, RAND's Logistics Department could have easily staffed several university-level departments in operations research. During the 1950s, RAND had been in the forefront of new developments in mathematical programming, Monte Carlo simulation, and discrete-event simulation, top-ics in which George had limited experience prior to joining RAND. In December1963, Murray Geisler, then chair of the Logistics Department, asked George to referee a RAND report on statistical aspects of simulation. After reading the review, Geisler encouraged George to examine the many outstanding statistical issues in simulation. During the next seven years at RAND, George wrote a collection of publications with a focus on estimating the statistical accuracy of simulation results." (Advancing the Frontiers of Simulation: A Festschrift in Honor of George Samuel Fishman. Alexopoulos, Goldsman & Wilson). Regarding Estimating Sample Size, "An algorithm for automatically estimating and collecting the sample size required for statistical precision in a computer simulation experiment while the simulation is running. The algorithm, which would be incorporated directly into the computer routines, would relieve an investigator of the burden of first estimating the variance of the sample mean from a data sample obtained from a trial run, then estimating the sample size necessary for the specified confidence interval, and finally collecting that many more observations in a successive simulation run. The underlying probability model is autoregressive: it would depend on an autoregressive representation of the sequence that considers each observation as a linear combination of past observations plus an uncorrelated random residual. This approach need not require more than four or five autocovariance computations to estimate the variance of the sample mean. A flowchart is included to aid in building the technique into simulation programs." (The Rand Corporation). HBS 68940. $500.…