Given a system that exhibits oscillating behavior for some set of parameters, the central question of robustness is how far the parameters may be changed, such that the qualitative behavior does not change.
The robustness of a model is determined by answering the question how far the parameters of the model could be perturbed so that the qualitative behavior of the system does not change.
Once a relationship between a CPP and CQA is demonstrated, the variable that demonstrated impact on a CQA should be further evaluated with the goal of understanding how much variability can be induced (i.e., how far the parameter range can be) in relation to other parameters as they collectively deliver acceptable product quality.
We next investigated in how far the kinetic parameters can influence the saturation concentration and the maximal output value at saturation (Additional file 3, Fig. S8 and Additional file 3, Fig. S9).
As opposed to classical Bayesian OED strategies which select the experiment that most reduces the uncertainty in parameter estimation, itself quantified by the variance or the entropy of the posterior parameter distribution, we formulate the problem in a decision-theoretic framework where we wish to minimize an error function quantifying how far the estimated parameters are from the true ones.
We also explored in how far varying the parameters within the ranges that likely are encountered in the habitat of giant squid, i.e., bioluminescent intensity from 10 to 10quanta/s and nearest neighbor distances of 0.45-0.65 m, influenced estimated visual performance.
A form of local sensitivity analysis, which reveals how far each individual parameter may be perturbed from calibrated baseline values before scientifically significant deviations in simulation behavior occur, was performed.
Equip T 3 with a distribution whose trace in each orthant is a normalized Gaussian distribution, centered at identical distance from the origin, and truncated at 0. By symmetry the Fréchet mean is at the origin, and one can ask how far the location parameter m of one component can be perturbed without affecting the mean.
"location parameter; how far away do I live from my workplace". .
"location parameter; how far away do I live from my workplace".
