Empirical mode modeling

Abstract Data-driven, model-free analytics are natural choices for discovery and forecasting of complex, nonlinear systems. Methods that operate in the system state-space require either an explicit multidimensional state-space, or, one approximated from available observations. Since observational data frequently sampled with noise, it is possible noise can corrupt representation degrading analytical performance. Here, we evaluate synthesis empirical mode decomposition dynamic modeling, which term to increase information content representations presence noise. Evaluation a mathematical, and, ecologically important geophysical application across three different suggests modeling may be useful technique data-driven, model-free, analysis