Bias correction for the estimation of sensitivity indices based on random balance designs

This paper deals with the random balance design method (RBD) and its hybrid approach, RBD-FAST. Both these global sensitivity analysis methods originate from Fourier amplitude sensitivity test (FAST) and are consequently faced with the main problems inherent to discrete harmonic analysis. As some authors pointed out in these methods, certain estimates of sensitivity indices are systematically over-evaluated. This positive bias has recently been identified in the RBD method by Xu and Gertner [1]. Following their work, we propose a bias correction method for first-order sensitivity indices estimates in RBD. We then extend the correction method to the sensitivity indices of any order in RBD-FAST. At last, we suggest an efficient strategy to estimate all the first and second-order sensitivity indices using RBD-FAST.