Package: ddsPLS 1.2.1

Overview | Tuning parameters | Quality descriptors | Construction and prediction errors | Bootstrap versions of the $R^2$ and the $Q^2$ | Automatic tunning of the parameters | Nota bene | Download and load the package | A Latent Variable Model | A high-dimensional structure | Correlation structure | A low correlated data-set | Evaluate the quality of the model | Compare with "non selection" ddsPLS model | The S3-method plot | Predicted versus observed values of the response | The criterion $\bar{R}{B}^2-\bar{Q}{B}^2$ | The $\bar{Q}_{B}^2$ metrics | The proportion of informational bootstrap models | Plot of the weights | Appendix | A.1 Definition of the "component-$r$ $\bar{Q}B^2$" denoted $\bar{Q}{B,r}^2$ | A.2 Function to simulate the high-dimensional data sets | A.3 Definitions of the explained variances... | A.3.1 Cumulated | $$\hat | A.3.2 Per component | A.3.3 Per response variable | A.3.4 Per response variable per component