2.6. Sensitivity analysis

Given the uncertainty of the parameters related to the transmission, testing and adherence, we employ the Latin Hypercube Sampling/Partial Rank Correlation Coefficient (LHS/PRCC) sensitivity analysis method to identify which parameters are more significant on the model outputs, such as cumulative cases and deaths. We generate 2000 samples using uniform distribution and used parameters related to Canada. The parameters studied and the sampling ranges are provided in Table S9. Parameters are defined as significant if their PRCC is, in absolute value, larger than 0.5.