Fig. 2
From: Exploring dynamic metabolomics data with multiway data analysis: a simulation study
Cross-validation performance of the Paralind(1,2,2) model and the 2-component CP model for the data generated by the linear open system with one source of induced variation and individual variation at level \(\beta =0.01\). The noise level is \(\eta =0.3\), and 20 TCS values are used in the boxplots. The missing data patterns are the same for both CP and Paralind(1,2,2), and thus the TCS is paired. The difference in TCS is statistically significant based on the paired t-test