A flexible and reusable framework for Bitcoin blockchain and price data

Abstract

zDriven by the potential for high returns and financial innovation, the cryptocurrency market is seeing a rapid increase in the number of investors and market value.Unlike other financial assets, cryptocurrencies have a transparent and accessible data structure that makes it possible to examine the relationships between price movements and on-chain metrics.This growing interest, as well as the transparent and accessible structure, is leading to a parallel increase in academic studies focused on analyzing the price dynamics of Bitcoin, the world's first widely accepted cryptocurrency, and the underlying factors by examining Bitcoin blockchain data and developing predictive models.The current literature generally relies on data collected at limited, fixed time intervals on a daily basis.However, the daily and fixed data sets of the cryptocurrency market in the literature become outdated over time and are not sufficient to analyze instantaneous price movements.This poses a significant limitation when analyzing cryptocurrency markets, which are highly volatile and require a focus on rapid price fluctuations and immediate market changes.For detailed analysis, investors and researchers increasingly require data at a higher frequency than daily.In this study, a framework is designed and developed that allows users to obtain Bitcoin blocks and price data for any desired time period and interval.In this way, a comprehensive, upto-date and flexible framework is offered that can serve various research areas, such as machine learning-based predictions, testing algorithmic trading strategies, investigating correlations between the blockchain and price fluctuations, performing market analysis or detecting anomalies.In addition, various frequency aggregations were defined using the framework developed in this study.In particular, a 10minute time interval was analyzed based on the average time it takes to generate a Bitcoin block.In this way, a direct and consistent relationship between block information and price movements could be established.