Search Results (93)

We propose a two-stage statistical learning framework to investigate how financial news headlines posted over weekends affect stock returns. In the first stage, Natural Language Processing (NLP) techniques are used to extract sentiment features from news headlines, including FinBERT sentiment scores and Impact Probabilities derived from Logistic Regression models (Binomial, Multinomial, and Bayesian). These Impact Probabilities estimate the likelihood that a given headline influences the stock’s opening price on the following trading day. In the second stage, we predict over-weekend log returns using various sets of covariates: sentiment-based features, traditional financial indicators (e.g., trading volumes, past returns), and headline counts. We evaluate multiple statistical learning algorithms—including Linear Regression, Polynomial Regression, Random Forests, and Support Vector Machines—using cross-validation and two performance metrics. Our framework is demonstrated using financial news from MarketWatch and stock data for Apple Inc. (AAPL) from 2014 to 2023. The results show that incorporating sentiment features, particularly Impact Probabilities, improves predictive accuracy. This approach offers a robust way to quantify and model the influence of qualitative financial information on stock performance, especially in contexts where markets are closed but news continues to develop. Full article

The stock market plays a crucial role in the financial system, with its price movements reflecting macroeconomic trends. Due to the influence of multifaceted factors such as policy shifts and corporate performance, stock prices exhibit nonlinearity, high noise, and non-stationarity, making them difficult to model accurately using a single approach. To enhance forecasting accuracy, this study proposes a hybrid forecasting framework that integrates wavelet denoising, multi-head attention-based BiLSTM, ARIMA, and XGBoost. Wavelet transform is first employed to enhance data quality. The multi-head attention BiLSTM captures nonlinear temporal dependencies, ARIMA models linear trends in residuals, and XGBoost improves the recognition of complex patterns. The final prediction is obtained by combining the outputs of all models through an inverse-error weighted ensemble strategy. Using the CSI 300 Index as an empirical case, we construct a multidimensional feature set including both market and technical indicators. Experimental results show that the proposed model clearly outperforms individual models in terms of RMSE, MAE, MAPE, and R². Ablation studies confirm the importance of each module in performance enhancement. The model also performs well on individual stock data (e.g., Fuyao Glass), demonstrating promising generalization ability. This research provides an effective solution for improving stock price forecasting accuracy and offers valuable insights for investment decision-making and market regulation. Full article

Globalization and the spread of technological innovations have made world markets and economies increasingly unified and conditioned by international trade, not only for sales markets but above all for the supply of raw materials necessary for the functioning of the production complex of each country. Alongside oil and gold, the main commodities traded include industrial metals, such as aluminum and copper, mineral products such as gas, electrical and electronic components, agricultural products, and precious metals. The conflict between Russia and Ukraine tested the unification of markets, given that these are countries with notable raw materials and are strongly dedicated to exports. This suggests that commodity prices were able to influence the stock markets, especially in the countries most closely linked to the two belligerents in terms of import-export. Given the importance of industrial metals in this period of energy transition, the aim of our study is to analyze whether Industrial Metals volatility affects G7 stock markets. To this end, the BEKK-GARCH model is used. The sample period spans from 3 January 2018 to 17 September 2024. The results show that lagged shocks and volatility significantly and positively influence the current conditional volatility of commodity and stock returns during all periods. In fact, past shocks inversely influence the current volatility of stock indices in periods when external events disrupt financial markets. The results show a non-linear and positive impact of commodity volatility on the implied volatility of the stock markets. The findings suggest that the war significantly affected stock prices and exacerbated volatility, so investors should diversify their portfolios to maximize returns and reduce risk differently in times of crisis, and a lack of diversification of raw materials is a risky factor for investors. Full article

Forecasting financial time series is challenging due to their intrinsic nonlinearity, high volatility, and complex dependencies across temporal scales. This study introduces MSGformer, a novel hybrid architecture that integrates multi-scale graph neural networks (MSGNet) with Transformer encoders to capture both local temporal fluctuations and long-term global trends in high-frequency financial data. The MSGNet module constructs multi-scale representations using adaptive graph convolutions and intra-sequence attention, while the Transformer component enhances long-range dependency modeling via multi-head self-attention. We evaluate MSGformer on minute-level stock index data from the Chinese A-share market, including CSI 300, SSE 50, CSI 500, and SSE Composite indices. Extensive experiments demonstrate that MSGformer significantly outperforms state-of-the-art baselines (e.g., Transformer, PatchTST, Autoformer) in terms of MAE, RMSE, MAPE, and R². The results confirm that the proposed hybrid model achieves superior prediction accuracy, robustness, and generalization across various forecasting horizons, providing an effective solution for real-world financial decision-making and risk assessment. Full article

This study investigates the causal impact of stock price crash risk on the cost of equity (COE) in China’s segmented A- and B-share markets with an emphasis on ownership structures and market regimes. Employing a bootstrap panel Granger causality framework, Markov-switching dynamic regression, and panel threshold regression models, the analysis reveals that heightened crash risk significantly increases COE, with the effects being more pronounced for A-shares because of domestic investors’ heightened risk sensitivity. This relationship further intensifies in bull markets, where investor optimism amplifies downside risk perceptions. Ownership segmentation plays a critical role, as foreign investors in B-shares exhibit weaker reliance on firm-level valuation metrics, favoring broader risk-diversification strategies. These findings offer actionable insights into corporate risk management, investor decision making, and policy formulation in segmented and emerging equity markets. Full article

Accurate stock price prediction requires the integration of heterogeneous data streams, yet conventional techniques struggle to simultaneously leverage fine-grained micro-stock features and broader macroeconomic indicators. To address this gap, we propose MambaLLM, a novel framework that fuses macro-index and micro-stock inputs through the synergistic use of state-space models (SSMs) and large language models (LLMs). Our two-branch architecture comprises (i) Micro-Stock Encoder, a Mamba-based temporal encoder for processing granular stock-level data (prices, volumes, and technical indicators), and (ii) Macro-Index Analyzer, an LLM module—employing DeepSeek R1 7B distillation—capable of interpreting market-level index trends (e.g., S&P 500) to produce textual summaries. These summaries are then distilled into compact embeddings via FinBERT. By merging these multi-scale representations through a concatenation mechanism and subsequently refining them with multi-layer perceptrons (MLPs), MambaLLM dynamically captures both asset-specific price behavior and systemic market fluctuations. Extensive experiments on six major U.S. stocks (AAPL, AMZN, MSFT, TSLA, GOOGL, and META) reveal that MambaLLM delivers up to a 28.50% reduction in RMSE compared with suboptimal models, surpassing traditional recurrent neural networks and MAMBA-based baselines under volatile market conditions. This marked performance gain highlights the framework’s unique ability to merge structured financial time series with semantically rich macroeconomic narratives. Altogether, our findings underscore the scalability and adaptability of MambaLLM, offering a powerful, next-generation tool for financial forecasting and risk management. Full article

Bitcoin’s role in global finance has rapidly expanded with increasing institutional participation, prompting new questions about its linkage to macroeconomic variables. This study thoughtfully integrates a Bayesian Markov Chain Monte Carlo (MCMC) covariate selection process within homogeneous and non-homogeneous Hidden Markov Models (HMMs) to analyze 16 macroeconomic and Bitcoin-specific factors from 2016 to 2024. The proposed method integrates likelihood penalties to refine variable selection and employs a rolling-window bootstrap procedure for 1-, 5-, and 30-step-ahead forecasting. Results indicate a fundamental shift: while early Bitcoin pricing was primarily driven by technical and supply-side factors (e.g., halving cycles, trading volume), later periods exhibit stronger ties to macroeconomic indicators such as exchange rates and major stock indices. Heightened volatility aligns with significant events—including regulatory changes and institutional announcements—underscoring Bitcoin’s evolving market structure. These findings demonstrate that integrating Bayesian MCMC within a regime-switching model provides robust insights into Bitcoin’s deepening connection with traditional financial forces. Full article

Reforestation is widely promoted as a nature-based solution for climate change, yet its unintended consequences, such as deforestation leakage, remain under-investigated. This study provides empirical evidence of reforestation-induced leakage in the Brazilian Amazon, using municipality-level panel data from 2000 to 2023 and spatial Durbin panel models to estimate both the magnitude and spatial reach of agricultural displacement. Despite the positive local effects of reforestation projects, we found a significant displacement of deforestation to the vicinity of municipalities. We estimated a statistically significant deforestation leakage effect of approximately 12% from the reforested area, due to the agricultural displacement of cattle ranching activities. Spatial spillovers are strongest within a 150 km radius and within two years after reforestation onset. Sensitivity tests using alternative spatial weight matrices, including distance decay and land rent-weighted specifications, confirm the robustness of these findings. Livestock intensification, proxied by cattle stocking rates, does not significantly mitigate displacement effects, challenging assumptions about land sparing benefits. These results suggest that current carbon market protocols (e.g., Verra, ART-TREES) may improve their leakage analysis to avoid under- or over-estimating net carbon benefits. Incorporating spatial econometric evidence into offset methodologies and reforestation planning can improve climate policy integrity and reduce unintended environmental trade-offs. Full article

This study presents an advanced adaptive trading framework that integrates Deep Reinforcement Learning (DRL) with the Iterative Model Combining Algorithm (IMCA) to overcome the critical limitations of static ensemble methods in global portfolio optimization. Using a diverse cross-market dataset of 39 stocks from the US, Australia, Europe, Thailand, and one cryptocurrency (BTC-USD), the research rigorously evaluates models’ adaptability under volatile market conditions. Volatile market conditions—such as COVID-19, SVB crisis, and the 2022 crypto crash—are captured via volatility metrics (e.g., drawdown), with DRL models like PPO/TD3 adapting through dynamic reward signals. This cross-asset integration is particularly critical, as it captures the complex dynamics and correlations between traditional financial markets and emerging digital assets. Although DRL models like PPO and TD3 outperform traditional strategies, they remain vulnerable to market drawdowns and high volatility. IMCA significantly surpasses these models, achieving the highest cumulative return of 29.52% and a superior Sharpe ratio of 0.829 by dynamically recalibrating model weights in response to real-time market dynamics. This study addresses a substantial research gap, highlighting the failure of traditional ensemble models—reliant on static weightings—to adapt to evolving financial conditions, resulting in suboptimal risk-adjusted returns. IMCA offers a dynamic, data-driven approach that continuously optimizes portfolio strategies across fluctuating market regimes, demonstrating its scalability and robustness across diverse asset classes and regional markets, and providing an empirical framework for adaptive portfolio management. Policy recommendations underscore the need for financial institutions to adopt AI-driven adaptive models like IMCA to enhance portfolio resilience, profitability, and responsiveness in uncertain markets. Full article

Stock return prediction is a pivotal yet intricate task in financial markets, challenged by volatility and multifaceted dependencies. This study proposes a hybrid model integrating long short-term memory (LSTM) networks and graph convolutional networks (GCNs) to enhance accuracy by capturing both temporal dynamics and spatial inter-stock relationships. Tested on the Dow Jones Industrial Average (DJIA), Shanghai Stock Exchange 50 (SSE50), and China Securities Index 100 (CSI 100), our LSTM-GCN model outperforms baselines—LSTM, GCN, RNN, GRU, BP, decision tree, and SVM—achieving the lowest mean squared error (e.g., 0.0055 on DJIA), mean absolute error, and highest R² values. This superior performance stems from the synergistic interaction of spatio-temporal features, offering a robust tool for investors and policymakers. Future enhancements could incorporate sentiment analysis and dynamic graph structures. Full article

This paper revisits the effects of financial market openness and stock market returns on economic development in the Gulf Cooperation Council countries over the period 1993–2022. We performed the panel stationarity test advanced that accommodates the presence of multiple structural breaks and exploits the cross-section variations. Empirical results from several panel tests provide strong support for the long-run positive effect of financial market openness on economic growth and a long-run negative association between stock market returns and growth. Findings of the robustness checks reveal that the effect of both financial market openness and stock market returns on economic growth differs across countries. Full article

Predicting forage biomass yield is critical in managing livestock since it impacts livestock stocking rates, hay procurement, and livestock marketing strategies. Only a few biomass yield prediction studies on pasture and rangeland exist despite the need. Therefore, this study focused on developing a biomass yield prediction methodology through remote sensing satellite imagery (multispectral bands) and climate data, employing open-source software technologies. Biomass ground truth data were obtained from local pastures, where Kentucky bluegrass is the predominant species among other forages. Remote sensing data included spatial bands (6), vegetation indices (30), and climate data (16). The top-ranked features (52 tested) from recursive feature elimination (RFE) were short-wave infrared 2, normalized difference moisture index, and average turf soil temperature in the machine learning (ML) model developed. The random forest (RF) model produced the highest accuracy ($R^2=0.83$) among others tested for biomass yield prediction. Applications of the developed methodology revealed that (i) the methodology applies to other unseen pasters ($R^2=0.79$), (ii) finer satellite spatial resolution (e.g., CubeSat; 3 m) better-predicted pasture biomass, and (iii) the methodology successfully developed for a combination of Kentucky bluegrass and other forages, extended to high-value alfalfa hay crop with excellent yield prediction accuracy ($R^2=0.95$). The developed methodology of RFE for feature selection and RF for biomass yield modeling is recommended for biomass and hay forage yield prediction. Full article

While there is a large body of literature on oil uncertainty-equity prices and/or returns nexus, an associated important question of how oil market uncertainty affects stock market bubbles remains unanswered. In this paper, we first use the Multi-Scale Log-Periodic Power Law Singularity Confidence Indicator (MS-LPPLS-CI) approach to detect both positive and negative bubbles in the short-, medium- and long-term stock markets of the G7 countries. While detecting major crashes and booms in the seven stock markets over the monthly period of February 1973 to May 2020, we also observe similar timing of strong (positive and negative) LPPLS-CIs across the G7, suggesting synchronized boom-bust cycles. Given this, we next apply dynamic heterogeneous coefficients panel databased regressions to analyze the predictive impact of a model-free robust metric of oil price uncertainty on the bubbles indicators. After controlling for the impacts of output growth, inflation, and monetary policy, we find that oil price uncertainty predicts a decrease in all the time scales and countries of the positive bubbles and increases strongly in the medium term for five countries (and weakly the short-term) negative LPPLS-CIs. The aggregate findings continue to hold with the inclusion of investor sentiment indicators. Our results have important implications for both investors and policymakers, as the higher (lower) oil price uncertainty can lead to a crash (recovery) in a bullish (bearish) market. Full article

Wood from reforestation gains market value due to its sustainable and legal origin. Planted forests in Brazil play a crucial role in economic, social and environmental aspects, with Eucalyptus and Pinus dominating the timber sector. However, non-majority species, such as those of the Khaya genus, have attracted great commercial interest due to the quality of their wood, being seen as an alternative to Brazilian mahogany. This study aimed to evaluate the biomass production of Khaya spp. stands and the nutrient uptake impacts in different harvesting scenarios. The research area is in Reserva Natural Vale (RNV) in Sooretama, Espírito Santo state, Brazil. The study was conducted 9.5 years after the planting of the Khaya spp. monoculture at a spacing of five m × five m, and the base fertilization consisted of 150 g of yoorin thermophosphate and 15 g of FTE BR 12 per seedling. The seedlings were of seminal origin, coming from different regions of Brazil and corresponding to three species: Kkaya grandifoliola C.DC (Belém-PA), Khaya ivorensis A. Chev. (Linhares-ES) and Khaya senegalensis A. Juss. (Poranguatu-GO). K. senegalensis exhibited the highest percentage of bark, while K. ivorensis was found to have the highest percentage of leaves. The biomass of the stems and branches did not vary by species. The relative biomass proportions had the following order: branches > stems > bark > leaves. The stocks of Ca and Mg were higher for K. grandifoliola, exceeding those for K. senegalensis (22.1%) for Ca and for K. ivorensis (42.3%) for Mg. The lowest nutrient uptake occurred in the scenario in which only the stem was removed, with averages of 44.17, 10.43, 21.93, 52.59 and 9.97 kg ha⁻¹ for N, P, K, Ca and Mg, respectively. Compared to total biomass harvesting, this represents a reduction in export levels by 91.34% for N, 79.31% for P, 94.66% for K, 94.29% for Ca and 93.28% for Mg. The nutrient uptake assessment demonstrated that more conservative harvest scenarios resulted in lower nutrient losses, indicating the importance of forest management practices that prioritize soil and nutrient conservation. In summary, the findings of this study provide a solid basis for the sustainable management of Khaya spp., highlighting implications for productivity and nutrient dynamics on a small or medium scale. Full article

Financial economists have long studied factors related to risk premiums, pricing biases, and diversification impediments. This study examines the relationship between a firm’s commitment to environmental, social, and governance principles (ESGs) and asset market returns. We incorporate an algorithmic protocol to identify three nonobservable but pervasive E, S, and G time-series factors to meet the study’s objectives. The novel factors were tested for information content by constructing a six-factor Fama and French model following the imposition of the isolation and disentanglement algorithm. Realizing that nonlinear relationships characterize models incorporating both observable and nonobservable factors, the Fama and French model statement was estimated using an enhanced shallow-learning neural network. Finally, as a post hoc measure, we integrated explainable AI (XAI) to simplify the machine learning outputs. Our study extends the literature on the disentanglement of investment factors across two dimensions. We first identify new time-series-based E, S, and G factors. Second, we demonstrate how machine learning can be used to model asset returns, considering the complex interconnectedness of sustainability factors. Our approach is further supported by comparing neural-network-estimated E, S, and G weights with London Stock Exchange ESG ratings. Full article