Time series forecasting of styrene price using a hybrid ARIMA and neural network model

Abstract

Every player in the market has a greater need to know about the smallest change in the market. Therefore, the ability to see what is ahead is a valuable advantage. The purpose of this research is to make an attempt to understand the behavioral patterns and try to find a new hybrid forecasting approach based on ARIMA-ANN for estimating styrene price. The time series analysis and forecasting is an essential tool which could be widely useful for finding the significant characteristics for making future decisions. In this study ARIMA, ANN and Hybrid ARIMA-ANN models were applied to evaluate the previous behavior of a time series data, in order to make interpretations about its future behavior for styrene price. Experimental results with real data sets show that the combined model can be most suitable to improve forecasting accurateness rather than traditional time series forecasting methodologies. As a subset of the literature, the small number of studies have been done to realize the new forecasting methods for forecasting styrene price.

Optimization of English Machine Translation by Deep Neural Network under Artificial Intelligence

“…The ultimate goal is to obtain the output sequence with the highest probability. The higher the probability of the sequence, the more reasonable the possibility of the sequence [ 14 – 16 ]. In the process of decoding, the decoder will search for the maximum conditional probability.…”