A note on (basic) Principal Components Analysis
- Autori: Muggeo VMR
- Anno di pubblicazione: 2020
- Tipologia: Contributo in atti di convegno pubblicato in volume
- OA Link: http://hdl.handle.net/10447/434258
Abstract
This paper provides some issues, known but somewhat little stressed, on using the conventional covariance or correlation matrix when performing the simple PCA. The paper also proposes a new simple alternative by providing some evidence, via real-data analysis and some simulation experiments, supporting the proposal