An Evolutionary Bootstrap Method for Selecting Dynamic Trading Strategies

摘要: This paper combines techniques drawn from the literature on evolutionary optimization algorithms along with bootstrap based statistical tests. Bootstrapping and cross validation are used as a general framework for estimating objectives out of sample by redrawing subsets training sample. Evolution is to search large space potential network architectures. The combination these two methods creates estimation selection procedure which aims find parsimonious structures generalize well. Examples given financial data showing how this compares more traditional model methods. methodology also allows objective functions than usual least squares since it can estimate in bias any function. Some will be compared estimates dynamic trading settings foreign exchange series.