Figure 9
(I) – Determination of optimized parameters and confidence intervals for the θ 8 , θ 31 fluxes from (A) “properly” corrupted and (B) “mistakenly” corrupted values of SER MIDs, and after the “elimination” of residuals corresponding to ALA MIDs (C) or SER MIDs (D) from the expression for the objective function. (II) – Stem plots of the variance-weighted individual residuals at the point of convergence. The individual residual values are represented by the leaves of the stem plots sorted in accordance with their enumeration provided by the model. Additional red horizontal lines represent the 0.95-quantile values of the standard normal distribution Ν(0, 1). The efflux values, estimated from an SLE, are usually rather close to the measured values; the last seven points, indicating the residuals corresponding to the efflux measurements, are close to zero. In the case of the PLE in which efflux measurements were generated independently for each SLE making up the PLE, this tendency was not observed.