Spline-DCS for Forecasting Trade Volume in High-Frequency Financial Data
Preprint
Repository URI
Repository DOI
Change log
Authors
Abstract
The Spline-DCS model is developed and applied to forecasting the high-frequency trade volume of selected equity and foreign currency exchange pairs. The cubic spline model of Harvey and Koopman (1993) is applied to capture intra-day periodic patterns. The model is robust to outliers as the dynamics of scale is driven by the score. The empirical application illustrates that Spline-DCS is a practical forecasting tool that is robust to the choice of sampling frequency or sampling period. The predictive performance of the model is compared with the state-of-the-art volume forecasting model, named the component-MEM, of Brownlees et al. (2011). The model can substantially outperform the component-MEM in minimizing common forecast error functions.
price impact