Maximum entropy spectral analysis for streamflow forecasting

Configurational entropy spectral analysis (CESAS) is developed with spectral power as a random variable for streamflow forecasting. It is found that the CESAS derived by maximizing the configurational entropy yields the same solution as by the Burg entropy spectral analysis (BESA). Comparison of forecasted streamflows by CESAS and BESA shows less than 0.001% difference between the two analyses and thus the two entropy spectral analyses are concluded to be identical. Thus, the Burg entropy spectral analysis and two configurational entropy spectral analyses form the maximum entropy spectral analysis. (C) 2015 Elsevier B.V. All rights reserved