Reconstruction Error: What You Lose with k Components
This lesson is part of the course
In this lesson
You rebuild the z-scores from one component (score × direction), measure the squared reconstruction error, express it as variance lost and compare k = 1 with k = 2 using MMULT.
Buy the course once for $9.99 to unlock every lesson and lab, the final exam and the certificate – lifetime access, no subscription.
Lessons in this course
- Standardising Variables: z-Scores before PCA Locked
- Covariance & Correlation Matrices with MMULT and TRANSPOSE Locked
- Power Iteration: the First Eigenvector with Repeated MMULT Locked
- Eigenvalue & Explained Variance: How Much Does PC1 Capture? Locked
- The Second Component: Deflation and Another Power Iteration Locked
- Scree Plot & Cumulative Variance: How Many Components? (with a Chart) Locked
- Component Scores & Loadings: Placing and Naming the Cities (with a Chart) Locked
- Reconstruction Error: What You Lose with k Components Locked