Relevance-weighted forecasting based on time-series decomposition

摘要

An input time-series is decomposed into a set of constituent frequencies. For each constituent frequency in a subset of the set of constituent frequencies, a corresponding forecasting model is selected in a subset from a set of forecasting models. From a set of component forecasts produced by the subset of forecasting models, a subset of component forecasts is selected. A component forecast in the subset of component forecasts is selected according to a component forecast selection condition. The subset of component forecasts is output to revise the forecast selection condition. A revised forecast selection condition increases a relevance of a future subset of component forecasts.