This document discusses using partial least squares regression (PLS) to address collinearity problems in multiple regression analysis and apply it to rainfall prediction using Global Circulation Model (GCM) data. PLS regression is an iterative method that can be used to solve collinearity issues unlike other methods that have closed-form solutions. The model was built using GCM and rainfall data from 1996-2000 and validated on 2001 data. PLS analysis over several iterations found a PRESS value of 5.714.118 and R^2 of 0.669 at the third iteration, indicating the model with 3 components could explain 66.9% of the variance in the dependent variable. The optimal PLS model based on GCM data