Package: LearnPCA 0.3.4

Audience | Why a Package? | About the Authors | Acknowledgements

Conceptual Introduction to PCA | PCA Results Illustrated, No Code, No Math | Refinements 1 | Refinements 2 | A Social Science Data Set | A Spectroscopic Data Set

Step 1. Centering the Data | Step 3. Data Reduction | Using prcomp | Using All the Data | What Else is in the PCA Results? | Scree Plot | Loading Plot | How Does prcomp Actually Work? | Step 4. Undoing the Scaling | Step 5. Undoing the Centering | Proof of Perfect Reconstruction | The More Components Used, the Better the Reconstruction

Introduction | A Small Data Set | Rotating the Axes | Scores | Loadings

Visualizing The Original Data Set | The First Principal Component | The Second Principal Component | The Third Principal Component | How PCA Changes the Data Cloud | Final Thoughts

Introduction | Matrix Decompositions | The SVD Decomposition | A Simple Implementation of SVD | Step 1 | Step 2 | Step 3 | Step 4 | Overall | Reporting | Comparison to the Answer from svd | Comparison to the Answer from prcomp | The Eigen Decomposition | A Simple Implementation of the Eigen Decomposition | The Relationship Between SVD and Eigen Decomposition | Singular Values vs Eigenvalues | A Note About the Meaning of "loadings" | Algorithmic Reality | The NIPALS Algorithm | A Simple Implementation of NIPALS | Checking Our work | Tying Things Together | References

prcomp | princomp | Covariance | Correlation | prcomp vs princomp | Compare the Scores | Reconstruct the Original Data | svd | Compare to the Scores from prcomp | Compare to the Loadings from prcomp | eigen | Compare to the Scores from princomp | Compare to the Loadings from princomp | Wrap Up

Warning When Analyzing Time Series and Spatial Data | Working with Nominal or Ordinal Response Data