Comprehensive Modeling Approaches for Forecasting Bitcoin Transaction Fees: A Comparative Study

2025-02-03

AI Summary

This paper addresses the short-term (24-hour/144-block) prediction of Bitcoin transaction fees. We propose a multi-source feature model integrating mempool state, network parameters, and historical fee data. Under a constrained dataset of 91 days, we systematically evaluate six forecasting approaches: SARIMAX, Prophet, Time2Vec, Time2Vec+Attention, SARIMAX-GBDT, and the Temporal Fusion Transformer (TFT). Results show that classical statistical models significantly outperform deep learning methods: SARIMAX achieves the highest accuracy on an independent test set, while Prophet yields the best performance under five-fold cross-validation; Time2Vec and TFT suffer from insufficient training samples. Based on these findings, we introduce, for the first time, a “data-scale-driven model selection principle,” elucidating the practical advantages and applicability boundaries of statistical models in small-sample time-series forecasting. This work provides a reproducible, methodology-oriented framework for on-chain fee modeling in blockchain systems.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Transaction fee prediction in Bitcoin's ecosystem represents a crucial challenge affecting both user costs and miner revenue optimization. This study presents a systematic evaluation of six predictive models for forecasting Bitcoin transaction fees across a 24-hour horizon (144 blocks): SARIMAX, Prophet, Time2Vec, Time2Vec with Attention, a Hybrid model combining SARIMAX with Gradient Boosting, and the Temporal Fusion Transformer (TFT). Our approach integrates comprehensive feature engineering spanning mempool metrics, network parameters, and historical fee patterns to capture the multifaceted dynamics of fee behavior. Through rigorous 5-fold cross-validation and independent testing, our analysis reveals that traditional statistical approaches outperform more complex deep learning architectures. The SARIMAX model achieves superior accuracy on the independent test set, while Prophet demonstrates strong performance during cross-validation. Notably, sophisticated deep learning models like Time2Vec and TFT show comparatively lower predictive power despite their architectural complexity. This performance disparity likely stems from the relatively constrained training dataset of 91 days, suggesting that deep learning models may achieve enhanced results with extended historical data. These findings offer significant practical implications for cryptocurrency stakeholders, providing empirically-validated guidance for fee-sensitive decision making while illuminating critical considerations in model selection based on data constraints. The study establishes a foundation for advanced fee prediction while highlighting the current advantages of traditional statistical methods in this domain.

Problem

• Bitcoin transaction
• fee prediction
• cost optimization

Innovation

• Bitcoin transaction fees
• SARIMAX model
• Deep learning vs statistical models