Fig. 3From: Sparse sliced inverse regression for high dimensional data analysisRiboflavin production data: left panel plot is the sufficient summary plot \((Y_{\text{train}}\) versus \(\widehat{\mathbf{v }}_1^{\top }{{\mathbf{X}}}_{\text{train}})\) for the 50 training samples; right panel is the sufficient summary plot \((Y_{\text{test}}\) versus \(\widehat{\mathbf{v }}_1^{\top }{{\mathbf{X}}}_{\text{test}})\) of the 21 test samplesBack to article page