RESUMO
Commodity futures constitute an attractive asset class for portfolio managers. Propelled by their low correlation with other assets, commodities begin gaining popularity among investors, as they allow to capture diversification benefits. This comprehensive study examines the time and frequency spillovers between the Economic Policy Uncertainty [1] and a broad set of commodities encompassing ferrous, non-ferrous, and precious metals, food, and energy commodities over a period from December 1997 to April 2022, which includes various political, economic and health crises. The novelty of this research lies in its extensive temporal and categorical coverage, providing an understanding of how different types of commodities respond to various crises. Furthermore, our study breaks new ground by employing wavelet analysis to gain detailed insights in both time and frequency domains in the financial time series of interest, providing a deeper understanding of the co-movements and lead-lag relationships. Specifically, we introduce the Cross Wavelet Transform (XWT) and Wavelet Coherence (WTC) analysis. Our findings demonstrate that not all crises uniformly impact commodities. Notably, during the global financial crisis and the COVID-19 pandemic, co-movements between commodities became significantly stronger. These results highlight the heterogeneity within the commodity asset class, where individual commodities exhibit diverse underlying dynamics. Importantly, the proposed methodology facilitates the extraction of robust results even when dealing with nonlinearities and nonstationary time series data. Consequently, our work offers valuable insights for policymakers (including regulatory bodies), investors, and fund managers.
RESUMO
Automatic text summarization tools have a great impact on many fields, such as medicine, law, and scientific research in general. As information overload increases, automatic summaries allow handling the growing volume of documents, usually by assigning weights to the extracted phrases based on their significance in the expected summary. Obtaining the main contents of any given document in less time than it would take to do that manually is still an issue of interest. In this article, a new method is presented that allows automatically generating extractive summaries from documents by adequately weighting sentence scoring features using Particle Swarm Optimization. The key feature of the proposed method is the identification of those features that are closest to the criterion used by the individual when summarizing. The proposed method combines a binary representation and a continuous one, using an original variation of the technique developed by the authors of this paper. Our paper shows that using user labeled information in the training set helps to find better metrics and weights. The empirical results yield an improved accuracy compared to previous methods used in this field.