Analysis of simulation with common random numbers: a note on Heikes et al. (1976)

ABSTRACT

If different random number streams are used then traditional Analysis of Variance can be applied. However, the standard errors of the estimated effects should account for heterogeneity of variance. If common random numbers are used then the quite complicated procedure proposed by Heikes et al. in 1976, may be replaced by the simple Student t-statistic combined with the so-called Bonferroni - inequality. The statistical techniques are illustrated with a simple case study.

REFERENCES

• 1: Graybill, F. A., Introduction to Linear Statistical Models. McGraw-Hill, New York, Vol. 1 1961.
• 2: Heikes, R. G., D. C. Montgomery and R. L. Rardin, Using common random numbers in simulation experiments - an approach to statistical analysis. Simulation, 27, no. 3, Sept. 1976, pp. 81--85.
• 3: Kleijnen, J. P. C., Comparing means and variances of two simulations. Simulation, 26, no. 3, March 1976, pp. 87--88.
• 4: Kleijnen, J. P. C., Regression Analysis for Simulation Practitioners. FEW 85, Dept. of Business and Economics, Katholieke Hogeschool, Tilburg (Neth.), 1979.
• 5: Miller, R. G., Simultaneous Statistical Inference. McGraw-Hill, New York, 1966.