21.4.3.10. Discrete Latin Hypercube Design

The original Latin Hypercube sampling is developed as a variance reduction technique or as a screening technique. The basic of Latin Hypercube sampling is a full stratification of sampled distribution with a random selection inside each stratum. The sample values are randomly shuffled among different variables. The original Latin Hypercube sampling has been widely used in the deterministic simulation for computer experimentation. However, we use the Latin Hypercube design as a space filling sampling for constructing Meta-Model. Thus, we modify it as maximum entropy samples in the latticed space, while maintaining the strength-I orthogonal characteristic of Latin Hypercube design. Figure 21.143 shows a discrete Latin Hypercube design for \(N=6\) and \(k=2\).

../_images/image180.png

Figure 21.143 A Discrete Latin Hypercube sample with \(N=6\) and \(k=2\)

Reference

  1. Michael, S., 1987, “Large Sample Properties of Simulations using Latin Hypercube Sampling”, Technometrics, Vol. 29, No.2, pp.143 ~ 151

  2. McKay, M.D., Conover,W.J. and Beckman, R.J., 1979, “A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output From a Computer Code”,Technometrics, Vol. 21, pp. 239 ~ 245.