MODELING HIGH-FREQUENCY FOREIGN EXCHANGE DATA DYNAMICS
This paper shows that high-frequency, irregularly spaced, foreign exchange (FX) data can generate nonnormality, conditional heteroskedasticity, and leptokurtosis when aggregated into fixed-interval calendar time, even when these features are absent in the original DGP. Furthermore, we introduce a new approach to modeling these high-frequency irregularly spaced data based on the Poisson regression model. The new model is called the autoregressive conditional intensity model and it has the advantage of being simple and of maintaining the calendar timescale. To illustrate the virtues of this approach, we examine a classical issue in FX microstructure: the variation in information content as a function of fluctuations in the intensity of activity levels.
Key Words: Time Aggregation; Irregularly Spaced High-Frequency Data; Dependent Point Process.
c1 The paper has benefited from valuable comments made by two anonymous referees and an associate editor. We thank Rob Engle, Tom Rothenberg, Jim Stock, and seminar participants at the European University Institute, Harvard University, the Midwestern Econometrics Group, University of California at Berkeley, University of California at Davis, and University of California at Riverside for useful suggestions. All errors remain our responsibility. Address correspondence to: Òscar Jordà, Department of Economics, University of California, Davis, One Shields Avenue, Davis, CA 95616-8578, USA; e-mail: [email protected].