Economic and financial decision-making may cause a significant impact on government, society, and industries. Due to the increasing volume of data, decision science has become an interdisciplinary field of study, supported by efficient methods and models of data analysis. Our contributions lie exactly in the intersection of signal processing, tensorial algebra, and decision science. More precisely, we introduce a novel approach in which the data taken into account in the decision process is modeled as a tensor. Moreover, we apply adaptive prediction methods for aggregating the decision tensor. Results provided by numerical experiments with both synthetic and actual data attest to the efficacy of our proposal in better supporting economic and financial decisions.