Wavelet packet transforms analysis applied to carbon prices

This paper deals with carbon price variations using a multi time scale decomposition based on the theory of wavelets. Our approach is based on wavelet packet transforms. This original approach enables us to identify that the periods which contribute the most to EUA spot, EUA futures, and CER futures price variations are February-April 2008, October-November 2008, and the recent 2009-2011 business cycle which correspond to major institutional uncertainties and changes in macroeconomic fundamentals. This wavelet decomposition therefore provides additional evidence on the drivers of carbon prices being institutional events and economic activity.