I have a few questions regarding how specifically principal components are formed:
- What is the relevance of the magnitudes of the covariance when it comes to eigendecomposition?
- How does eigendecomposition form the principal components?
- How are the different covariances weighted?
- Can the weightings be vaguely predicted from the covariance matrix?
