Recent Developments: Chaos and Neural Networks
Imad A. Moosa · Palgrave Macmillan UK eBooks · 2000
This chapter deals with some recent developments that are not specifically related to exchange rate forecasting. Rather, they are new developments that originated in the biological and physical sciences and then were imported to economics and finance. These two topics are dealt with in one chapter not because they are inherently related, but because they are new developments that have some implications for exchange rate forecasting. Indeed, the two topics have completely contrasting implications for exchange rate forecasting. Chaos theory tells us that if the process generating exchange rates is chaotic then it is deterministic but unpredictable. Conversely, enthusiastic proponents of neural networks as a forecasting technique claim that it is superior to conventional techniques when it comes to forecasting economic and financial variables, including exchange rates. The conclusions derived from the discussion of these two topics and the implications for exchange rate forecasting are completely the opposite. We start with a discussion of chaos theory. Recent years have witnessed an explosion of interest in this subject by scholars from various disciplines. The interest has even engulfed the public media. This may sound really strange, given that this topic is extremely demanding when it comes to the level of mathematics required to handle it. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.