Complex mathematical simulations are performed to understand the various phenomena in physics, chemistry, etc. Usually these models consist of several input variables and sensitivity analysis explains the variation of computer code output due to the changes in the input values. We have used the software package, Gaussian Emulation Machine for sensitivity analysis (GEM-SA), (Kennedy and O'Hagan, 2006) for performing sensitivity analysis and chosen Rosenbrock function as a test problem. In this paper we have studied the effect of different available software routines for Latin hypercube designs (LHD) on values of the variance-based sensitive indices, total sensitive indices and roughness parameters for the test problem. We have used MATLAB, R software and GEM-SA modules for LHD.
Rao, D. M., & Konda, J. (2010). Computation of sensitivity indices with different routines for Latin hypercube designs. In Procedia - Social and Behavioral Sciences (Vol. 2, pp. 7730–7731). https://doi.org/10.1016/j.sbspro.2010.05.201