Supplementary Materials for: A Hybrid SVR-Based Framework for Cryptocurrency Price Forecasting and Strategy Backtesting (Version 3)

This dataset (Version 3) accompanies the manuscript titled "A Hybrid SVR-Based Framework for Cryptocurrency Price Forecasting and Strategy Backtesting." Compared to earlier versions, this release includes: Prediction results from additional machine learning models: Random Forest (RF) and Long Short-Term Memory (LSTM), as requested by peer reviewers; Updated tables for four major cryptocurrencies (BTC, ETH, XRP, LTC); Selected Jupyter notebooks focusing on the prediction component of each model (SVR, RF, LSTM); Clarified and structured folder organization. Please note: Only partial code is provided in this release (prediction only). Full implementation details are available upon request from the corresponding author. The data can be used to reproduce the main results and model comparisons presented in the manuscript. For citation, please use the canonical DOI: 10.5281/zenodo.15261521 — this links to all versions of the dataset.