ABSTRACT
Among all computational intelligence (CI) tools, the artificial neural network (ANN) is the most widely acceptable tool for economists and finance people, even though its history is much shorter than that of fuzzy logic so far as application to economics and finance are concerned. The earliest application of ANNs appeared in 1988 by Halbert White (White, 1988). Since then we have witnessed an exponential growth in the number of applications. ANN is probably the only CI tool which drew serious attention from econometricians and on which a lot of theoretical studies have been done. Both White (1992) and Refenes and Zapranis (1999) gave rigorous mathematical or statistical treatments of ANNs and hence they have established ANNs with a sound foundation in the econometric field. Nowadays, it has already become an essential chapter of any textbook in econometrics and, in particular, financial econometrics and financial time series. A great number of textbooks or volumes specially edited for economists and finance people are available, to name a few: Trippi and Turban (1993), Azoff (1994), Baestaens, Van Den Bergh, and Wood (1994), Refenes (1995), Gately (1996), Zirilli (1996), and Shadbolt and Taylor (2002). Its significance to finance people can also be seen from the establishment of the journal Neurove$t (now Computational Intelligence in Finance) in 1993.
