Using signal processing techniques to characterise change in the dynamics of foreign exchange markets following political and economic events

摘要

The paper proposes a method for investigating the dynamics of foreign exchange markets using power spectral analysis combined with filter design. This has allowed a more precise description and understanding of the way in which the market character has changed following certain key political, economic and technological events such as the growth of electronic trading. Results following investigation of the last ten years' worth of hourly trading for the Euro//sup S/ are presented. Some interesting outcomes and implications of the technique for supporting trading decision-making are discussed.