A SPECTRAL DECOMPOSITION APPROACH TO SEPARATING INDEPENDENT FACTORS: THE CASE OF FOREIGN EXCHANGE RATES
Sorin-Manuel Delureanu Student · RePEc: Research Papers in Economics · 2015
In this paper, Independent Component Analysis (ICA) is addressed for revealing fundamental factors behind 6 parallel series of foreign exchange rates. ICA is belonging to the blind source separation (BSS) area of research. Each measured signal is considered a mixture of several distinct underlying factors. Separating such fundamental causal factors is very important for multivariate financial time series analysis, in order to explain past co-movements and to predict future evolutions.