Testing for Endogeneity of Irregular Sampling Schemes

In the context of high-frequency financial data it is often assumed that sampling times are exogenous. This entails that financial asset prices, sampled on a grid of trade instants, are independent from the sampling times. We derive statistical tests capable of determining whether or not, and to what extent, this hypothesis is rejected by the data. We test for sampling time endogeneity in relation to both the efficient and the noise components of the observed price. Using a vast dataset of financial asset prices we give empirical evidence that the efficient component of the observed price process does not show a dependence with trade arrival times of the kind that may jeopardize well-known results on convergence of power variations. In addition, we provide empirical evidence that the assumption of independence between market microstructure noise and trading instants is not supported by the data.