The Unscrambler如何做PCA.ppt

PCAPrincipal Component Analysis This presentation covers the following Why PCA Principles behind PCA Interpretation of PCA plots in The Unscrambler Note: This presentation was made for versions of PowerPoint 2000 and earlier If you would like the version of this presentation for PowerPoint XP, please contact RanjitV@CAMO.com Principal Component Analysis (PCA) Why PCA? A typical data analysis situation 12 Jams samples were made from berries plucked in various cultivars and seasonal times. Several parameters (sensory measurements) were measured on each sample. Data set Raspberry Jams Sample comparison according to 1 variable: Redness Sample comparison according to 2 variables: Redness and colour Sample comparison according to 3 variables: Redness, colour and R. Smell Sample comparison according to all 12 variables: multivariate model (PCA) Sample comparison according to all 12 variables: multivariate model (PCA) Sample comparison according to all 12 variables: multivariate model (PCA) Multivariate Analysis Applies in any situation where several variables were measured Does not study the multiple variables separately; it utilizes all variables into 1 model Handles missing values Fully exploits the information contained in your data Principal Component Analysis (PCA) Principles behind PCA The principles of Principal Component Analysis (PCA) The principles of Principal Component Analysis (PCA) The principles of Principal Component Analysis (PCA) PCA- Scores Plot PCA- Scores Plot PCA- Scores Plot PCA- Scores Plot PCA – Loadings Plot PCA – Loadings Plot PCA – Loadings Plot PCA – Loadings Plot PCA – Loadings Plot PCA – Bi Plots PCA – Bi Plots Principal Component Analysis (PCA) Interpretation of Plots PCA Overview Loadings Plot(top right plot) Loadings Plot(top right plot) Scores Plot(top left plot) Scores Plot(top left plot) Bi-PlotScores and Loadings on same plot For more details, please feel free to contact us Ranjit Viswanathan Manager – Sales (Asia Pacific) CA