Econ Papers Daily / 期刊论文
加密货币市场无套利检验
Tests for no-arbitrage in cryptocurrency markets
摘要
This article investigates the efficiency of the cryptocurrency markets by testing the no-arbitrage hypothesis in a general framework for cryptocurrencies traded on multiple exchange platforms. We test the moment conditions implied by the no-arbitrage hypothesis by applying a simulated method of moments and a general specification test. For each of the five cryptocurrencies considered, the test is applied globally using the full sample and locally using a rolling window approach. Globally, the results show that the null hypothesis of no-arbitrage is rejected for Bitcoin and Ethereum, but is not rejected for Binance Coin and for the two stablecoins, USD Coin and Tether. Locally, the null hypothesis is never rejected for USD Coin and is only rejected 5.5% of the time for Tether. In comparison, the test rejects the null hypothesis frequently for the three studied native cryptocurrencies (39% of the time for Bitcoin, 26.5% for Ethereum, and 22% for Binance Coin).
中文摘要
本文在一个适用于在多个交易平台交易的加密货币的一般框架下，通过检验无套利假说来研究加密货币市场的效率。我们通过应用模拟矩方法和一般设定检验，检验无套利假说所隐含的矩条件。对于所考察的五种加密货币中的每一种，该检验均使用全样本进行全局应用，并使用滚动窗口方法进行局部应用。在全局层面，结果表明，无套利的原假设对于比特币和以太坊被拒绝，但对于币安币以及两种稳定币USD Coin和Tether未被拒绝。在局部层面，该原假设对于USD Coin从未被拒绝，对于Tether仅在5.5%的情况下被拒绝。相比之下，该检验对于所研究的三种原生加密货币频繁拒绝原假设（比特币为39%的时间，以太坊为26.5%，币安币为22%）。