We introduce a new nonparametric method to measure microstructure noise, the deviation of the observed asset prices from the fundamental values caused by market imperfections. Using high-frequency data, we provide joint estimators of arbitrary finite moments of microstructure noise, which could be serially dependent and nonstationary. We characterize the limit distributions of the proposed estimators and construct robust confidence intervals under infill asymptotics. We further demonstrate a consistency property of our new estimators without any specification on the data frequencies. As an economic application, we propose two liquidity measures that gauge the instantaneous and average bid-ask spread with potentially autocorrelated order flows, and such measures can be interpreted as an intermediary’s inventory risks to meet liquidity demand. Statistical applications include several model-free tests for the intraday patterns and the zero autocorrelations hypotheses of microstructure noise.