We investigate interaction effects in cryptocurrency markets by constructing and evaluating double-sorted portfolios based on 40 different characteristics. Using a dataset of over 500 major coins and tokens from 2017 to 2023, we identify numerous significant interactions. The most pronounced effects arise from the interplay of liquidity, risk, and past return measures. An out-of-sample long-short strategy that selects the top and bottom interactions achieves a Sharpe ratio exceeding 1. However, network graph analysis and additional tests reveal that low liquidity, which raises transaction costs, can dampen trading activity and contribute to the persistence of these anomalies.

中文翻译：

---

加密货币回报中的横截面交互作用


我们通过根据 40 种不同的特征构建和评估双重排序的投资组合来研究加密货币市场的交互效应。使用 2017 年至 2023 年 500 多种主要硬币和代币的数据集，我们确定了许多重要的相互作用。最明显的影响来自流动性、风险和过去回报指标的相互作用。选择顶部和底部交互作用的样本外多空策略可实现超过 1 的夏普比率。然而，网络图分析和其他测试表明，提高交易成本的低流动性会抑制交易活动并导致这些异常现象的持续存在。