A state space approach to estimating the integrated variance under the existence of market microstructure noise

This article develops a state space method for estimating the integrated variance under the existence of market microstructure noise (MMN). Our method is based on a state space representation of the noisecontaminated RV (NCRV), namely, the realized variance (RV) calculated with observed prices contaminated by MMNs. The main idea of our method is to filter out the bias component, which we call the microstructure noise (MN) component, from the NCRV using the Kalman filter. We apply the proposed method to yen/dollar exchange rate data.We find that about half of the variation in NCRV is because of the MN component. The proposed method can serve as a convenient way to estimate a general class of continuous-time stochastic volatility models under the existence of MMN.