Efficient generation of exponential and normal deviates
Herman Rubin, Brad C. Johnson · Journal of Statistical Computation and Simulation · 2006
We present efficient procedures for generating random exponential and normal deviates based on the acceptance-complement method [Kronmal, R.A. and Peterson, A.V., 1981, Journal of the American Statistical Association, 76, 446–451.]. We provide comparisons with the corresponding Ziggurat procedures proposed by Marsaglia and Tsang [Marsaglia, G. and Tsang, W.W., 1984, SIAM Journal on Scientific and Statistical Computing, 5, 349–359; Marsaglia, G. and Tsang, W.W., 2000, Journal of Statistical Software, 5(8), 1–7.]. The proposed procedures maintain good precision over the entire support of the respective densities and are very easy to set up and implement. The proposed exponential procedure compares favourably with the Ziggurat procedure in terms of speed, running up to 24% faster on some platform/compiler/uniform generator combinations tested.