Sensitivity analysis is considered as an important part of evaluating the performance of mathematical or numerical models. One-factor-at-a-time (OAT) and differential methods are among the most popular sensitivity analysis (SA) schemes employed in the literature. The two major limitations of the above methods are lack of addressing the correlation between model factors and being a local method. Given these limitations, its extensive use among modelers raises concern over the credibility of the associated sensitivity analyses.This paper proposes proof of the inefficiency of the aforementioned methods drawing from experimental designs, and provides a novel technique based on principal component analysis (PCA) to address the issue of the correlation among input factors. In addition, proper guidelines are suggested to handle other conditions.
Daneshbod, Y., & Abedini, M. J. (2016). Proposing a global sensitivity analysis method for linear models in the presence of correlation among input variables. Scientia Iranica, 23(2), 399-406. doi: 10.24200/sci.2016.2126
MLA
Y. Daneshbod; M. J. Abedini. "Proposing a global sensitivity analysis method for linear models in the presence of correlation among input variables". Scientia Iranica, 23, 2, 2016, 399-406. doi: 10.24200/sci.2016.2126
HARVARD
Daneshbod, Y., Abedini, M. J. (2016). 'Proposing a global sensitivity analysis method for linear models in the presence of correlation among input variables', Scientia Iranica, 23(2), pp. 399-406. doi: 10.24200/sci.2016.2126
VANCOUVER
Daneshbod, Y., Abedini, M. J. Proposing a global sensitivity analysis method for linear models in the presence of correlation among input variables. Scientia Iranica, 2016; 23(2): 399-406. doi: 10.24200/sci.2016.2126
)