Supporting data for the manuscript "TempoStackNet: An Advanced Stacking-Based Network for Enhanced Multi-Step Cryptocurrencies Prediction Using Temporal Attention Mechanisms". Mechanisms".

The data was downloaded from Yahoo Finance and divided into three main datasets.The first one is the USD price of Bitcoin (btc_data.csv), ranging from February 24, 2022, 00:00 to February 21, 2024, 23:00, comprising 17,165 data points with a time interval of 1 hour.The second dataset is the USD price of Ethereum (en_data.csv), spanning from February 24, 2022, 00:00 to February 21, 2024, 23:00, consisting of 12,342 data points with a time interval of 1 hour.The third dataset is the S&P 500 index (sp500_data.csv), covering the period from February 24, 2022, 00:00 to February 21, 2024, 23:00, containing 11,221 data points with a time interval of 1 hour.Regarding data partitioning, 80% of the data is allocated for training, while the remaining 20% is for testing. We employ a multi-step forecasting approach with a time step set to 32 steps.