文章基本信息

标题：COMPARISON OF TWO APPROACH OF NEURAL NETWORK FOR FORECASTING OF OIL FUTURES PRICES
本地全文：下载
作者：Mohsen Mehrara ; Ali Moeini ; Mehdi Ahrari 等
期刊名称：Asian Journal of Business and Management Sciences
电子版ISSN：2047-2528
出版年度：2012
卷号：1
期号：03
页码：54-64
出版社：Society for Business Research Promotion
摘要：Neural networks have the advantage of simulating the non-linear models when little a priori knowledge of the structure of problem exist or the number of immeasurable input variables are great and system has a chaotic characteristic. Among different methods, MLFF neural network with back-propagation learning algorithm and GMDH neural network with Genetic algorithm (GA) learning are used to predict oil futures price based on the NYMEX database. This paper uses moving average crossover inputs. The results indicates (1) there is short-term dependence in oil futures price movements and the Exponential Moving Average (EMA) has better result than Simple Moving Average (SMA) and (2) by means of GMDH approach, prediction accuracy, power tracking and profitability in comparison to MLFF neural networks, can be improved. Full Text