A novel system for fast and accurate decisions of gold-stock markets in the short-term prediction

The prediction of gold-stock price indices is considered a great challenge for traders, who try to analyze the historical developments and movements of gold-stock indices. Especially in the short-term prediction, that required a fast accurate decision in order to predict the next price after 5, 10, 15, 30 and 60 min. In this paper, a novel system is proposed to predict gold-stock prices in the short term for gold exchange markets. It is based on the wrapper selection method, artificial neural network and genetic algorithm which consist of three steps. The first step is to determine the number of the effective previous gold-stock indices and selecting the most significant technical parameters. Then, the prediction model construction will be trained by World gold council database. Finally, the next gold-stock price for each case of the short-term periods will be predicted. The results show that the proposed system achieves fast decisions with higher prediction accuracy and nuance rate between the actual and predicted values of the next gold-stock price after 5 min in the short term, which consumed 18 ms for the run time of the price prediction process.