During the unfolding of a crisis, it is crucial to forecast its severity at an early stage , yet access to reliable data is often challenging early on. The wisdom of crowds has been effective at forecasting in similar scenarios. We investigated whether the initial regional social media reaction to the emerging COVID-19 pandemic in three critically affected countries has significant relations with their observed mortality a month later. We obtained COVID-19 related regionally geolocated tweets from Italian, Spanish, and United States regions. We quantified the predictive power of the wisdom of the crowds using correlations and regressions of geolocated Tweet Intensity (TI) during the initial social media attention peak versus the cumulative number of deaths a month ahead. We found that the intensity of initial COVID-19 related tweet attention at the beginning of the pandemic across Italian, Spanish, and United States regions is significantly related (p < 0.001) to the extent to which these regions had been affected by the pandemic a month later. This association is most striking in Italy as when at its peak of TI in late February 2020 only two of its regions had reported mortality. The collective wisdom of the crowds at early stages of the pandemic, when information on the number of infections was not broadly available, strikingly predicted the extent of mortality reflecting the regional severity of the pandemic almost a month later. Our findings could underpin the creation of real-time novelty detection systems aimed at early reporting of the severity of crises impacting a territory leading to early activation of control measures at a stage when available data is extremely limited.
Logistic regression (LR) and support vector machine algorithms, together with linear and nonlinear deep neural networks (DNNs), are applied to lending data in order to replicate lender acceptance of loans and predict the likelihood of default of issued loans. A two-phase model is proposed; the first phase predicts loan rejection, while the second one predicts default risk for approved loans. LR was found to be the best performer for the first phase, with test set recall macro score of 77.4 % . DNNs were applied to the second phase only, where they achieved best performance, with test set recall score of 72 % , for defaults. This shows that artificial intelligence can improve current credit risk models reducing the default risk of issued loans by as much as 70 % . The models were also applied to loans taken for small businesses alone. The first phase of the model performs significantly better when trained on the whole dataset. Instead, the second phase performs significantly better when trained on the small business subset. This suggests a potential discrepancy between how these loans are screened and how they should be analysed in terms of default prediction.
We introduce the first end-to-end Deep Reinforcement Learning (DRL) based framework for active high frequency trading. We train DRL agents to trade one unit of Intel Corporation stock by employing the Proximal Policy Optimization algorithm. The training is performed on three contiguous months of high frequency Limit Order Book data, of which the last month constitutes the validation data. In order to maximise the signal to noise ratio in the training data, we compose the latter by only selecting training samples with largest price changes. The test is then carried out on the following month of data. Hyperparameters are tuned using the Sequential Model Based Optimization technique. We consider three different state characterizations, which differ in their LOB-based meta-features. Analysing the agents' performances on test data, we argue that the agents are able to create a dynamic representation of the underlying environment. They identify occasional regularities present in the data and exploit them to create long-term profitable trading strategies. Indeed, agents learn trading strategies able to produce stable positive returns in spite of the highly stochastic and non-stationary environment.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.