1 Analysis of complex computer simulators

5 Computer simulators of real world processes are often computationally expensive 6 and require many inputs. The problem of the computational expense can be handled 7 using emulation technology; however, highly-multidimensional input spaces may re8 quire more simulator runs to train and validate the emulator. We aim to reduce the 9 dimensionality of the problem by screening the simulator’s inputs for non-linear effects 10 on the output rather than distinguishing between negligible and active effects. Our 11 proposed method is built upon the elementary effects method for screening (Morris, 12 1991) and utilises a threshold value to separate the inputs with linear and non-linear 13 effects. The technique is simple to implement and acts in a sequential way in order to 14 keep the number of simulator runs down to a minimum, whilst identifying the inputs 15 that have non-linear effects. The algorithm is applied on a set of simulated examples 16 and a rabies disease simulator where we observe run savings ranging between 28% and 17 63% compared to the batch elementary effects method. Supplementary materials for 18 this paper are available online. 19