Research Center for Endogenous Resource Valorization

Funder

Organizational Unit

Publications

From wars to waves: geopolitical risks and environmental investment behaviour

Publication . Gabriel, Vítor; Dionísio, Andreia; Almeida, Dora; Ferreira, Paulo

This study investigates the impact of geopolitical risk (GPR) on sustainable investments, focusing on five global environmental indices and two global GPR indices. Using Corrected Dynamic Conditional Correlation Generalised Autoregressive Conditional Heteroskedasticity (cDCC-GARCH) model and Diebold and Yilmaz’s spillover analysis, we use daily data from January 2009 to October 2022, covering various market phases, including the European sovereign debt crisis, the COVID-19 pandemic, and the war in Ukraine. Results from the cDCC-GARCH model reveal high dynamic conditional correlations. During periods of high volatility, environmental indices displayed simultaneous and more intense responses, limiting investment diversification alternatives when considering only the environmental side. Diebold and Yilmaz’s static analysis demonstrates that environmental segments are more influenced by systemic shocks than specific causes, with GPR’s influence proving relatively weak. In the dynamic analysis, the spillover effects of GPR in environmental segments intensified during the pandemic crisis and the invasion of Ukraine, affecting market conditions.

2024-03-05Journal article

Open access

When two banks fall, how do markets react?

Publication . Almeida, Dora; Dionísio, Andreia; Ferreira, Paulo Jorge Silveira

The most recent fall of the Silicon Valley (SVB) and Credit Suisse (CS) banks increased the fear of a worldwide banking crisis. We analyse the impacts of their fall on five financial indices. We apply detrended fluctuation analysis, static and with sliding windows. We find a higher impact of the SVB fall on the efficiency dynamic of the studied indices, which revealed fluctuating efficiency and a loss of efficiency during the period of the falls. The fall of both banks contributed to some persistence in stock indices returns. The Nasdaq and STOXX Europe 600 Banks are the most and the least efficient indices, respectively. Despite the apparent evidence of inefficiency, it might not necessarily mean a capacity for abnormal profits.

2023-12-15Journal article

Open access

Impact of the COVID-19 Pandemic on Cryptocurrency Markets: A DCCA Analysis

Publication . Almeida, Dora; Dionísio, Andreia; Ferreira, Paulo Jorge Silveira; Isabel Vieira

Extraordinary events, regardless of their financial or non-financial nature, are a great challenge for financial stability. This study examines the impact of one such occurrence—the COVID19 pandemic—on cryptocurrency markets. A detrended cross-correlation analysis was performed to evaluate how the links between 16 cryptocurrencies were changed by this event. Cross-correlation coefficients that were calculated before and after the onset of the pandemic were compared, and the statistical significance of their variation was assessed. The analysis results show that the markets of the assessed cryptocurrencies became more integrated. There is also evidence to suggest that the pandemic crisis promoted contagion, mainly across short timescales (with a few exceptions of non-contagion across long timescales). We conclude that, in spite of the distinct characteristics of cryptocurrencies, those in our sample offered no protection against the financial turbulence provoked by the COVID-19 pandemic, and thus, our study provided yet another example of ‘correlations breakdown’ in times of crisis.

2023-05-24Journal article

Open access

Evidence of Intraday Multifractality in European Stock Markets during the Recent Coronavirus (COVID-19) Outbreak

Publication . Aslam, Faheem; Mohti, Wahbeeah; Ferreira, Paulo

2020-05-26Journal article

Open access

Hyperspectral Image Classification: An Analysis Employing CNN, LSTM, Transformer, and Attention Mechanism

Publication . Viel, Felipe; Renato Cotrim Maciel; Seman, Laio Oriel; Zeferino, Cesar Albenes; Bezerra, Eduardo; LEITHARDT, VALDERI

Hyperspectral images contain tens to hundreds of bands, implying a high spectral resolution. This high spectral resolution allows for obtaining a precise signature of structures and compounds that make up the captured scene. Among the types of processing that may be applied to Hyperspectral Images, classification using machine learning models stands out. The classification process is one of the most relevant steps for this type of image. It can extract information using spatial and spectral information and spatial-spectral fusion. Artificial Neural Network models have been gaining prominence among existing classification techniques. They can be applied to data with one, two, or three dimensions. Given the above, this work evaluates Convolutional Neural Network models with one, two, and three dimensions to identify the impact of classifying Hyperspectral Images with different types of convolution. We also expand the comparison to Recurrent Neural Network models, Attention Mechanism, and the Transformer architecture. Furthermore, a novelty pre-processing method is proposed for the classification process to avoid generating data leaks between training, validation, and testing data. The results demonstrated that using 1 Dimension Convolutional Neural Network (1D-CNN), Long Short-Term Memory (LSTM), and Transformer architectures reduces memory consumption and sample processing time and maintain a satisfactory classification performance up to 99% accuracy on larger datasets. In addition, the Transfomer architecture can approach the 2D-CNN and 3D-CNN architectures in accuracy using only spectral information. The results also show that using two or three dimensions convolution layers improves accuracy at the cost of greater memory consumption and processing time per sample. Furthermore, the pre-processing methodology guarantees the disassociation of training and testing data.

2016Journal article

Open access