NIPALS

NIPALS (Nonlinear Iterative Partial Least Squares) is a classic algorithm for Principal Component Analysis. It iteratively extracts principal components one at a time by alternating between estimating scores and loadings through least-squares updates. After a component is extracted, the data matrix is deflated and the next principal component is calculated. It can handle missing data by skipping them during its iterative updates. NIPALS is widely used in chemometrics and spectroscopy, where missing values frequently occur.