Developed multiple-layer perceptron neural network based on developed search and rescue optimizer to predict iron ore price volatility: A case study

Abstract

In economic investment, the role of forecasting is very important because in an economic project, the investor must carefully examine the dimensions of the work such that one of the most important and perhaps the main factor of a future investor and an economic enterprise is the work done by Costs and revenues are determined. Due to the fact that the volatility of iron ore price is affected by various factors, so it is not possible to determine a simple and general function to predict its price. There are several methods for predicting price, but the most appropriate of these is a method that examines variables in a nonlinear and dynamic manner that is closer to reality. Therefore, in this research, an improved and optimized neural network is proposed to facilitate this task. The idea is to employ a developed version of Search and Rescue optimization algorithm to enhance the training ability of the neural network to present an efficient forecasting tool for iron ore price volatilities. Different variables are used for the method verification and its results are compared with basic neural network, particle swarm optimization-based, Intelligent Integrated Optimizer, Genetic Neural Network to show its superiority. Simulation results demonstrate that by the proposed method has a satisfying and better fitting with the data compared with the other methods.