The aim of this study was to analyze and predict the prices of precious metals—palladium, platinum, rhodium, and gold—for the period from October 16 to November 27, 2024.
To achieve this goal, historical price data were utilized, and modern machine learning methods, specifically neural networks, were applied in combination with basic statistical analyses. By comparing predicted and actual values, it was possible to assess the accuracy of individual models and identify trends in price fluctuations. The findings showed that neural networks are effective at capturing price trends, particularly for metals with higher stability, such as rhodium. However, for palladium and platinum, the models tended to underestimate growth phases and smooth out extreme fluctuations. The results also highlighted the importance of geopolitical events and economic shocks, which significantly impact the volatility of precious metal markets. These factors were not fully included in the modeling, leading to reduced prediction accuracy in some cases. The contribution of this study lies in providing a comprehensive perspective on the price trends of key precious metals essential for the automotive industry. The findings may prove valuable not only for car manufacturers but also for investors and other stakeholders. A limitation of the research is the lack of deeper integration of macroeconomic and geopolitical variables, which presents an opportunity for further studies in this field.