Accurate gold price forecasting is essential for informed financial decision-making, as gold is sensitive to economic, political, and social factors. This study presents a hybrid framework for multivariate gold price prediction that integrates classical econometric modelling, traditional machine learning, modern deep learning methods, and their combinations. The framework incorporates financial, macroeconomic, and sentiment indicators, allowing it to capture complex temporal patterns and cross-variable relationships over time. Empirical validation on an eleven-year dataset (2014–2024) demonstrates the framework effectiveness across diverse market conditions. Results show that advanced supervised techniques outperform traditional econometric models under dynamic market environment. Key advantages of the framework include its ability to handle multiple data types, apply a structured variable selection process, employ diverse model families, and support model hybridisation and meta-modelling, providing practical guidance for investors, institutions, and policymakers.

Volatility forecasting plays a crucial role in financial markets, portfolio management, and risk control. Classical econometric models such as GARCH, ARIMA, and HAR-RV are widely used but face limitations in capturing the nonlinear and regime-dependent dynamics of financial volatility. This study compares traditional econometric models (HAR-RV, ARIMA, GARCH) with deep learning (DL) architectures (LSTM, CNN-LSTM, PatchTST-lite, and Vanilla Transformer) in forecasting realized variance (RV) for major U.S. equity indices (S&P 500, NASDAQ 100, and the Dow Jones Industrial Average) over the period 2000–2025. RV is used as the dependent variable because it is a standard model-free proxy for market volatility. Forecast accuracy is evaluated across forecast horizons of h = 1, 5, 22 days using QLIKE, RMSE, and MAE, along with Diebold–Mariano (DM) significance tests and overfitting diagnostics. Results show that Transformer-based models achieve the lowest errors and strongest generalization, particularly at short horizons and during volatile periods. Overall, the findings highlight the growing advantage of AI-driven models in delivering stable and economically meaningful volatility forecasts, supporting more effective portfolio allocation and risk management—especially in environments marked by rapid market shifts and structural breaks.

The investment potential of the cryptocurrency market is examined through an analysis of the factors that lead to sharp fluctuations and a forecast of its volatility is made with the use of the CRIX index. This index is dynamic and its structure changes every three months, which provides detailed information on the volatility of the cryptocurrency market. The CRIX index includes the most liquid cryptocurrencies, which makes it a representative indicator of the cryptocurrency market as well as a reliable indicator when it comes to making market forecasts in the future. The Autoregressive Integrated Moving Average (ARIMA) model of forecasting dynamic rows used in the research is a barometer of the cryptocurrency market. The research follows the changes in the monthly price of the CRIX index from January 2015 to January 2019.

Table of Content - Article Detail - Proceedings of the International Conference on Economics, Finance & Business, Venice

Accurate prediction of gold prices is essential for informed financial decision-making, given their sensitivity to economic, political, and social factors. This study proposes a new hybrid framework for forecasting gold prices, combining traditional financial modelling, classical machine learning, and advanced deep learning methods, including long short-term memory networks and their variations. The framework integrates financial, macroeconomic, and sentiment indicators through feature fusion, capturing complex temporal dynamics and cross‑variable dependencies to improve prediction accuracy. The experimental evaluation spans a ten-year period (2014–2024), allowing for a robust assessment of framework performance in forecasting gold futures under varying market conditions. Comparative analysis of classical econometric and modern machine‑learning models shows that advanced methods achieve higher forecasting accuracy and remain robust despite data variability.

The aim of this study is to analyze how fear affects investor behavior in decentralized financial (DeFi) markets. The focus is on the relationship between the Fear and Greed Index (F&G) and the market prices of governance tokens from major DeFi platforms, including PancakeSwap, Balancer, Compound, SushiSwap, Uniswap, Aave, and MakerDAO. Each of these tokens is a native asset linked to a specific DeFi protocol and, beyond their tradability, they function as governance instruments, making them particularly sensitive to shifts in market sentiment and investor expectations. To examine how investor fear and greed impact token price dynamics, we apply a combination of econometric and behavioral approaches. Our findings suggest that while investor sentiment may influence volatility and regime shifts, its direct predictive power over daily returns of governance tokens remains weak.