3
PCA and EOFs
¥Principal component analysis (Hotelling, 1933)
¥Empirical orthogonal functions (Lorenz, 1956)
¥Other names too
¥Reduces dimensionality by finding linear combinations of a large set of variables that successively maximise variance
¥Limitations
–Can be more difficult to interpret than using a subset of the original variables, but typically not for space-time data
– Linearity. Non-linear versions exist – not discussed here.  
–Uses only covariances, not higher-order moments – see independent component analysis (ICA)
¥