Tbe empirical results discussed in tbis article show that frequently a smaller quantity of training data will produce a better-performing backpropagation neural network model of a financial time series. Other problematic issues related to tbe development of the best possible neural network model—such as selection of input variables or selection of the neural network training algorithm and design of the neural network architecture—are not discussed. The prudent financial time series neural network developer will realize that these other factors will affect the neural network model's performance and should utilize existing guidelines [10, 25, 27, 32, 39] to solve these other issues.