What Is the Appropriate Range for R2 and Q2 Intercepts in PLS-DA
In Partial Least Squares Discriminant Analysis (PLS-DA), R2 is used to measure how well the model fits the data, whereas Q2 assesses the model's predictive capability. Ideally, R2 should approach 1, and Q2 should also be a positive value, with values closer to 1 indicating better performance.
During cross-validation, examining the intercepts of the regression lines for R2 and Q2 can provide insights into potential overfitting tendencies of the model.
1. An intercept for R2 (R2Y and R2X intercepts) that is near 0 is preferable, suggesting that the model has not captured random noise but instead accurately reflects the relationships within the data.
2. A Q2 intercept should be less than 0.05 or even lower, which indicates that there is minimal randomness in prediction, demonstrating the model's strong predictive power.
MtoZ Biolabs, an integrated chromatography and mass spectrometry (MS) services provider.
Related Services
How to order?
