2024· International Journal of Commerce, Finance and Digital Economy· Vol 7, pp. 1-13· 0 citations
TL;DR
Experimental results demonstrate that hybrid forecasting models outperform conventional statistical approaches by effectively capturing nonlinear temporal patterns and improving prediction accuracy under volatile market conditions.
Abstract
Commodity price forecasting plays a crucial role in international trade, agriculture, investment, and economic decision-making. However, accurate prediction remains challenging due to market volatility, economic uncertainty, geopolitical events, climate change, and supply chain disruptions. This study proposes a comprehensive time series forecasting framework that integrates classical statistical models, including ARIMA, SARIMA, and Exponential Smoothing, with machine learning and deep learning techniques such as Random Forest Regression, Support Vector Regression, Gradient Boosting, and Long Short-Term Memory (LSTM). The framework incorporates data preprocessing, feature engineering, trend decomposition, stationarity testing, model optimization, and rolling window validation to improve forecasting performance. Model evaluation is conducted using MAE, RMSE, MAPE, SMAPE, and R² metrics. Experimental results demonstrate that hybrid forecasting models outperform conventional statistical approaches by effectively capturing nonlinear temporal patterns and improving prediction accuracy under volatile market conditions. The proposed framework offers a scalable, interpretable, and adaptable solution for commodity price forecasting, supporting strategic decision-making, risk management, inventory optimization, and investment planning across diverse commodity sectors.
The findings demonstrate that rigorous leakage-free validation is essential for reliable forecasting research and that, for monthly Robusta coffee prices, increased model complexity does not necessarily yield superior predictive performance.
Dler H Kadir, D. Khalil, Azhin M. Khudhur· Forecasting· 0 citations
Accurate stock price forecasting remains a challenging problem due to the noisy, nonlinear, and non-stationary characteristics of financial time series. Although recent advances in deep learning have improved predictive capabilities, most prior studies evaluate forecasting models primarily using statistical error metrics, with limited consideration of their practical value in trading and investment decision-making. This creates a gap between predictive performance and real-world economic utility. This study proposes a decision-oriented evaluation framework for multi-step stock price forecasting that jointly assesses predictive accuracy and trading profitability within a unified experimental setting. Using data from 91 ASX 100 stocks after data cleaning, with a testing period spanning 2019–2020, several deep learning architectures, including Multi-Layer Perceptron (MLP), Gated Recurrent Unit (GRU), Seq2Seq, and attention-based sequence models, are systematically compared under identical training and trading conditions. The results show that the Seq2Seq model achieved the best overall performance, obtaining the lowest average MAPE of 0.0293 and the highest ROI of 23.2%, while the attention-based model achieved a similar MAPE of 0.0294 but a lower ROI of 12.4%. Although differences in forecasting accuracy were relatively small, the Seq2Seq model achieved the highest observed trading profitability and generated a higher observed return than a passive market benchmark under the proposed evaluation framework. These findings suggest that evaluation based solely on prediction accuracy may not fully capture the practical value of forecasting models.
C. Hargreaves, Hieu Le Trung· Algorithms· 0 citations
Demand forecasting plays a critical role in retail by influencing inventory management, supply chain efficiency, and customer satisfaction. Traditional statistical methods, while effective in stable environments, often fail to capture the complex and nonlinear patterns of modern retail data influenced by seasonality, promotions, and changing consumer behavior. With the advancement of Artificial Intelligence (AI), particularly machine learning and deep learning, demand forecasting has shifted toward more adaptive and accurate models. This paper presents a comprehensive study of AI-based demand forecasting models tailored for the retail sector. It examines various techniques, including regression models, decision trees, ensemble methods, neural networks, and deep learning approaches such as Long Short-Term Memory (LSTM) networks. The study highlights the ability of these models to process large, high-dimensional datasets, including both structured and unstructured data such as historical sales, pricing, weather, and consumer sentiment. The proposed framework integrates feature engineering, model training, and performance evaluation using metrics like Mean Absolute Percentage Error (MAPE) and Root Mean Squared Error (RMSE). Results indicate that hybrid models combining statistical and AI approaches improve forecasting accuracy. Overall, AI-based models significantly enhance prediction accuracy, helping retailers optimize inventory, reduce costs, and improve decision-making.
Moussa Camara· International Journal of App...· 0 citations
Stock market volatility is a key factor influencing investment decisions, portfolio optimization, risk management, derivative pricing, and financial planning. Traditional statistical models often struggle to capture the nonlinear, dynamic, and high-dimensional nature of modern financial markets. Recent advances in Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) provide more accurate and adaptive approaches for volatility prediction. This paper proposes a comprehensive predictive modeling framework that integrates financial data preprocessing, feature engineering, machine learning algorithms, and deep learning models for intelligent stock market volatility forecasting. The framework incorporates historical stock prices, technical indicators, trading volume, economic variables, financial news, and market sentiment to improve predictive performance. Advanced algorithms such as Random Forest, Support Vector Machine, Gradient Boosting, Extreme Gradient Boosting, Artificial Neural Networks (ANN), Long Short-Term Memory (LSTM), and Gated Recurrent Units (GRU) are employed to capture complex market patterns and temporal dependencies. Model performance is evaluated using metrics including Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), R², precision, recall, F1-score, and prediction accuracy. The proposed framework emphasizes scalability, computational efficiency, and adaptability to rapidly changing market conditions, supporting intelligent portfolio management and financial risk assessment. Results indicate that AI-driven predictive models outperform traditional statistical approaches in stock market volatility forecasting, providing a reliable foundation for data-driven investment strategies and next-generation intelligent financial analytics systems.
Seshagiri N, Mahabala H. N.· International Journal of Com...· 0 citations
This paper intends to predict the prices of commodities in metals, energy, agriculture, and other industrial products sectors. This study covers a large span of about ten and a half years with 3484 daily observations starting from 7 April 2016 to 7 April 2026. The commodities taken into account for this study are: gold, silver, copper, Brent oil, natural gas, coal, corn, wheat, soybean, aluminum, nickel, and lead. On the methodological front, this paper is a little different as it conducts a comparative analysis of out-of-sample forecasting capability of four models, namely, a naïve one-day-lag (random walk) benchmark, a linear autoregressive model, a bagged decision-tree ensemble model, and a gradient-boosted decision-tree ensemble model. Forecast accuracy is measured in terms of error-based and directional forecasting metrics, while Diebold-Mariano tests, sector-level comparisons, volatility-regime analysis, and robustness checks serve as evaluation tools for qualitative assessments. The results highlight that forecasting performance differs not only with the time horizon but also with the specific sectors. A linear autoregressive model ranks as the most reliable and regularly successful among the four models under comparison. A bagged decision-tree ensemble model and a gradient-boosted decision-tree ensemble model, on the other hand, as machine learning tree-based models, may only offer fleeting and unstable gains. The further the forecast extends in time (from 1 day to 5 and 10 days), the less accurate it gets. Energy commodities are the hardest to predict, while industrial commodities seem to be relatively more predictable.
Ahmet Kaya, Nazan Güngör Karyagdi, Mehmet Ozcalici et al.· Journal of Risk and Financia...· 0 citations
Price instability and fluctuations of basic commodities in East Java pose significant challenges that affect household purchasing power and complicate regional inflation control. This study aims to develop and evaluate a forecasting model for selected food commodity prices using a decomposition-based Hybrid ARIMA–LSTM framework. Daily time-series data from 2015–2024 are decomposed into trend, seasonal, and residual components. The linear trend component is modeled using ARIMA, while separate Long Short-Term Memory (LSTM) networks are employed to model the seasonal and residual components over a 30-day forecasting horizon. The framework is evaluated using medium rice, bulk cooking oil, and purebred chicken eggs across Surabaya City, Madiun City, and Banyuwangi Regency. Experimental results indicate that the decomposition-based parallel architecture improves short-term forecasting performance compared with standalone models by reducing error accumulation between linear and non-linear components. Quantitatively, ARIMA consistently achieves a coefficient of determination (R²) above 0.96 when modeling trend components, while LSTM effectively captures seasonal patterns. However, residual-component forecasting exhibits substantially lower performance in several datasets, reflecting the difficulty of modeling irregular stochastic variations using a purely time-series approach. Overall, the proposed framework demonstrates the potential of decomposition-based hybrid forecasting for selected regional food commodity prices and provides a foundation for future forecasting-oriented decision-support applications.
Muhammad Zaki Nawwafi, H. Wahanani, Andreas Nugroho Sihananto· bit-Tech· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
