Application of projection-pursuit principal component analysis method to climate studies

Projection-pursuit (PP) prinicipal component analysis (PCA) is a new statistical method that can deal with high-dimensional problems. In this paper, we show how this method can be applied to analyse regional monthly sea surface temperature and rainfall. Comparisons are made with results derrived from the traditional empirical orthogonal funtion (EOF) method. The effect of simulated outliers in the original data on the results is then examined for these two methods. The PP-PCA method is shown to be much more robust than the EOF method. This suggests that the former should be considered as an alternative in many of the climate sutdies.