Stock price prediction has been a subject of significant interest in the financial mathematics field. Recently, interest in natural language processing models has increased, and among them, transformer models, such as BERT and FinBERT, are attracting attention. This study uses a mathematical framework to investigate the effects of human sentiment on stock movements, especially in text data. In particular, FinBERT, a domain-specific language model based on BERT tailored for financial language, was employed for the sentiment analysis on the financial texts to extract sentiment information. In this study, we use “summary” text data extracted from The New York Times, representing concise summaries of news articles. Accordingly, we apply FinBERT to the summary text data to calculate sentiment scores. In addition, we employ the LSTM (Long short-term memory) methodology, one of the machine learning models, for stock price prediction using sentiment scores. Furthermore, the LSTM model was trained by stock price data and the estimated sentiment scores. We compared the predictive power of LSTM models with and without sentiment analysis based on error measures such as MSE, RMSE, and MAE. The empirical results demonstrated that including sentiment scores through the LSTM model led to improved prediction accuracy for all three measures. These findings indicate the significance of incorporating news sentiment into stock price predictions, shedding light on the potential impact of psychological factors on financial markets. By using the FinBERT transformer model, this study aimed to investigate the interplay between sentiment and stock price predictions, contributing to a deeper understanding of mathematical-based sentiment analysis in finance and its role in enhancing forecasting in financial mathematics. Furthermore, we show that using summary data instead of entire news articles is a useful strategy for mathematical-based sentiment analysis.
Since the breakdown of the Bretton Woods system in the early 1970s, the foreign exchange (FX) market has become an important focus of both academic and practical research. There are many reasons why FX is important, but one of most important aspects is the determination of foreign investment values. Therefore, FX serves as the backbone of international investments and global trading. Additionally, because fluctuations in FX affect the value of imported and exported goods and services, such fluctuations have an important impact on the economic competitiveness of multinational corporations and countries. Therefore, the volatility of FX rates is a major concern for scholars and practitioners. Forecasting FX volatility is a crucial financial problem that is attracting significant attention based on its diverse implications. Recently, various deep learning models based on artificial neural networks (ANNs) have been widely employed in finance and economics, particularly for forecasting volatility. The main goal of this study was to predict FX volatility effectively using ANN models. To this end, we propose a hybrid model that combines the long short‐term memory (LSTM) and autoencoder models. These deep learning models are known to perform well in time‐series prediction for forecasting FX volatility. Therefore, we expect that our approach will be suitable for FX volatility prediction because it combines the merits of these two models. Methodologically, we employ the Foreign Exchange Volatility Index (FXVIX) as a measure of FX volatility. In particular, the three major FXVIX indices (EUVIX, BPVIX, and JYVIX) from 2010 to 2019 are considered, and we predict future prices using the proposed hybrid model. Our hybrid model utilizes an LSTM model as an encoder and decoder inside an autoencoder network. Additionally, we investigate FXVIX indices through subperiod analysis to examine how the proposed model’s forecasting performance is influenced by data distributions and outliers. Based on the empirical results, we can conclude that the proposed hybrid method, which we call the autoencoder‐LSTM model, outperforms the traditional LSTM method. Additionally, the ability to learn the magnitude of data spread and singularities determines the accuracy of predictions made using deep learning models. In summary, this study established that FX volatility can be accurately predicted using a combination of deep learning models. Our findings have important implications for practitioners. Because forecasting volatility is an essential task for financial decision‐making, this study will enable traders and policymakers to hedge or invest efficiently and make policy decisions based on volatility forecasting.
Are green investments decoupled from the dirty investment such as the fossil fuel markets? We address this issue by extending the literature on environmental, social, and governance (ESG) assets by examining the dynamic relationship between fossil fuels and digital ESG assets proxied by green cryptocurrencies using the TVP-VAR(Time-varying parameter vector auto regression) spillover framework. Furthermore, we analyze the hedging attributes of green cryptocurrencies and fossil fuels in a minimum connectedness framework. The main findings are as follows: First, green cryptocurrencies are the main shock transmitters in all asset systems. Second, the dynamic connectedness between green cryptocurrencies and fossil fuels increased during the COVID-19 and Russia-Ukraine conflicts. Third, green cryptocurrencies have shown considerable hedging effectiveness against the fossil fuels. Our study has important implications for investors, regulators, and policy makers, such as shifting to green cryptocurrencies, regulation of carbon footprint, and promoting eco-friendly assets.
We investigate the relationship between crude oil prices and stock markets. Unlike prior studies, we use implied volatility indices and evaluate the change in the relationship between the volatility indices through a sub-period analysis. Specifically, we examine the causal relationships among the crude oil, S&P 500 index, and KOSPI 200 index volatilities by using the autoregressive distributed lag (ARDL) bounds and the Toda–Yamamoto Granger causality tests. In addition, a BEKK-GARCH model is employed to enhance the robustness of the causality test results. These experiments indicate that the OVX and VIX show bi-directional causality in the period that includes the shale gas revolution and no causality in the period that does not. Further, the OVX Granger causes the VKOSPI in the former period, but there is no causality between them in the latter period. Finally, we find strong unidirectional causality from the VIX to the VKOSPI in both sub-periods. These results have important implications for the analysis of portfolio risk management and for assisting energy policymakers and traders in making effective decisions and investments, respectively.
In this article, we investigate the market efficiency of global stock markets using the multifractal detrended fluctuation analysis methodology and analyze the results by dividing them into developed, emerging, and frontier groups. The static analysis results reveal that financially advanced countries, such as Switzerland, the UK, and the US, have more efficient stock markets than other countries. Rolling window analysis shows that global issues dominate the developed country group, while emerging markets are vulnerable to foreign capital movements and political risks. In the frontier group, intensive domestic market issues vary, making it difficult to distinguish similar dynamics. Our findings have important implications for international investors and policymakers. International investors can establish investment strategies based on the degree of market efficiency of individual stock markets. Policymakers in countries with significant fluctuations in market efficiency should consider implementing new regulations to enhance market efficiency. Overall, this study provides valuable insights into the market efficiency of global stock markets and highlights the need for careful consideration by international investors and policymakers.
We investigate the topology of sectoral returns in the US stock market using minimum spanning tree (MST) analysis. We examine four distinct time periods: the full period, the Global Financial Crisis (GFC), the COVID-19 pandemic, and the Russia-Ukraine war period. By comparing the static results across these periods, we identify differences in the network structure. Additionally, a rolling window analysis is conducted to explore the time-varying nature of the MST. We employ a TVP-VAR based connectedness framework to ensure a robust analysis of the sectoral return linkages. Our main findings are summarized as follows: First, the structure of the MST varies in different periods, with distinct crisis period structures. During the GFC, the industrial sector dominated clustering, whereas COVID-19 affected the financial, IT, and industrial sectors. The Russia-Ukraine war period showed clustering centered on materials, except in the industrial sector. These varying structures may explain the different characteristics of each crisis. Second, both static and rolling window analyses highlight the significance of the industrial sector in the US stock market. Third, the utilities sector exhibits the lowest centrality measures, indicating its minimal importance and lack of relationships with other industries. These findings provide valuable insights into the interrelationships among industries in the US stock market. Market participants can leverage these findings to enhance their understanding and improve their portfolio management. By utilizing this information, investors can develop optimal diversification strategies to maximize returns and minimize risk.