Scaling laws of Stablecoin Transactions: Evidence from USDT and USDC on the Ethereum blockchain

Battle Count: 82

The findings on heavy-tailed power-law scaling in stablecoin transaction values are relevant for understanding market microstructure and tail risk in DeFi and stablecoin-based trading. The distinction between EOA-driven and SC-driven transaction regimes could inform execution strategies and liquidity provision in Ethereum-based markets. However, the paper is primarily descriptive/foundational rather than directly proposing trading strategies. The scaling exponents (alpha ~1.4-1.8) are consistent with known financial market patterns, suggesting stablecoin markets share statistical properties with traditional markets, which is relevant for risk modeling.

The methodology involves standard statistical techniques: MLE for power-law fitting, KS distance minimization for threshold selection, and weighted composition analysis. The main computational challenge is processing ~370 million blockchain transactions. The analytical framework is well-established in econophysics literature. Requires familiarity with heavy-tailed distributions, maximum-likelihood estimation, and blockchain data structures. No deep learning or complex optimization is involved.

The paper provides detailed methodology including MLE formulas, KS distance minimization procedure, and sensitivity analysis parameters. Data is publicly available via XBlock-ETH dataset. Contract addresses for USDT and USDC are specified. Preprocessed data available from authors upon request. Computational facilities acknowledged. However, no code repository is explicitly linked.