Predictive Modeling of Bitcoin Halving Cycles Using AI-Based Trend Forecasting

This working paper explores the predictive power of artificial intelligence in modeling Bitcoin price behavior around halving events. Using historical data, the study applies Prophet (developed by Meta) and XGBoost to forecast post-halving trends and assess investment timing windows. Key findings show that halving cycles consistently precede 12–18 month bullish phases, with predictive indicators including halving dates, RSI, and trading volume. Benchmarked against expert forecasts (e.g., PlanB, Ark Invest), AI-based models demonstrated lower variance and more reliable long-term signals. Prepared by Hanzel Lacida (ORCID: 0009-0005-9770-1077), this research highlights the convergence of AI, data science, and fintech strategy. The paper does not provide financial advice and emphasizes replicability, with datasets and code to follow.