Toward a computable approach to the efficient market hypothesis: an application of genetic programming

A biological-search program, inspired by the genetic programming paradigm proposed by Koza (1992), has been cited effective in evaluating the unpredictability of efficient market hypothesis. Through this method, the concept of search programming is treated in way that both predictability and profitability are taken into account. Designed to adapt with the general style of evolutionary computation, the method is capable of beating 50% of random walk, which traditional linear models are incapable of achieving. The GP method was also evaluated through chaotic time series applications.