Deep Neural Network Architectures for Momentary Forecasting in Dry Bulk Markets: Robustness to the Impact of COVID-19

Abstract

The COVID-19 pandemic has severely affected various global markets, increasing the need for new forecasting models for the dry bulk market. Therefore, this study proposes deep neural network (abbreviated DNN) architectures to build a model for momentary forecasting that does not affect accuracy in the case of economic shocks (i.e., COVID-19) and elucidates the strategy for obtaining DNNs. First, since momentary and short-term forecastings are fundamentally different, they might use independent methods; as such, I apply DNN for the time series classification to momentary forecasting. Second, the proposed architecture is constructed by considering sparsity, because designing DNN architectures robust to any impacts is a type of overfitting prevention for deep neural networks. Finally, this study proposes indices for quantitatively evaluating the DNN architectures that represent the realized forecasting performance of various deep neural networks. Using these indices, I demonstrate that optimal architectures may need to have model sparsity in the DNN (i.e., sparsity independent of the input data). The importance of this issue has been demonstrated experimentally. As a result, the architectures achieved target performances of 88%, 91%, and 79% accuracy and had stability for Panamax, Supramax, and Capesize vessels, respectively from February 2016 to September 2021 (i.e., five years and eight months). It is difficult to identify a correlation between model performance and volatility. Furthermore, before and after the COVID-19 shock, the performance of the proposed models compared to the optimal one exceeds that of other four recent models, namely "Facebook Prophet," "DARTS," "SKTIME," and "AutoTS".

A Hybrid Deep Neural Approach for Segmenting the COVID Affection Area from the Lungs X-Ray Images

“…In deep neural networks, there are many hidden layers in-between, the output layer and input layers [ 11 ]. That hidden layer weighs the trained samples before entering the classification process [ 12 ].…”