Predicting the UK stock index after the outbreak of the Russia-Ukraine conflict

Accurate asset price forecasting is crucial for financial activities. With the continuous development of the Russia-Ukraine conflict, the global financial market has responded significantly and strongly to the geopolitical situation. As an international financial center, the UK market has also been impacted, with severe fluctuations in the UK stock index, potentially affecting investors. Therefore, this study selects the UK stock index as the research object, using three models to predict the UK market and evaluate model performance. The results indicate that SVM's fitting performance is inferior compared to that of the other two models during training. On test data, SVM exhibits poor predictive ability, while decision tree and random forest demonstrate high accuracy in their predictions. In particular, random forests provide a more precise simulation of price trends compared to decision trees, which closely aligns with actual value trends. This finding suggests that utilizing ensemble methods like random forest can enhance forecasting capabilities when analyzing stock market behavior.