Estimating High Frequency Foreign Exchange Rate Volatility with Nonparametric ARCH Models

High frequency foreign exchange rate (HFFX) series are analyzed on an operational time scale using models of the ARCH class. Comparison of the estimated conditional variances focuses on the asymmetry and persistence issue. Estimation results for para-metric models connrm standard results for HFFX series, namely high persistence and no signiicance of the asymmetry coeecient in an EGARCH model. To nd out whether these results are robust against alternative speciications, nonparametric models are estimated. Local linear estimation techniques are applied to a nonparametric ARCH model of order one (CHARN). The results show signiicant asymmetry of the volatility function. To allow for both exibility and persistence, a higher order multiplicative model is tted. The results show important asymmetries in volatility. In contrast to the EGARCH speciication, the news impact curves have diierent shapes for diierent lags and tend to increase slower at the boundaries.