A New Method for Analyzing Correlations Among Currency Exchange Rates

The correlation between different currency exchange rates has been studied for many years and a number of techniques have been developed. In this paper, we present a new algorithm to analyze the correlation between exchange rates based on biclustering. This algorithm is comprised of two parts. In the first part, the fast Hough transform is used to detect the lines in the exchange rate pair space. This phase is also called sub-biclustering and every line identified represents a sub-bicluster. In the second part, the sub-biclusters are combined based on comparison and merging. Experiment results show that this biclustering algorithm is very effective. The bicluster patterns are consistent with the underlying economic reasons.